Panel - Advancing Your Data Science Career During the Pandemic - #380

**Navigating Uncertainty: A Conversation on Career Transition and Personal Growth**

As we navigate this unprecedented time of uncertainty, it's natural to wonder how to proceed with our careers. For many of us, this is not a situation we've encountered before, and the lack of clear guidance can be unsettling. In a recent conversation, the topic of career transition came up, and one speaker shared their own experience of leaving the job market due to circumstances beyond their control.

"I really believe it," they said. "I just got off the job market because I was looking for a new job and really like, I had to make a lot of 'letting go' ego things." This sentiment resonated with many in the audience, who were grappling with similar feelings of uncertainty about their own career paths.

**The Value of Diversity in Career Transitions**

When it comes to transitioning into a new field, one speaker emphasized the importance of not under-selling oneself. "I would say your value is exponentially higher when you bring together all the marketing expertise and your data science knowledge," they noted. This point was echoed by others, who highlighted the need for individuals to showcase their unique strengths and perspectives in their career endeavors.

In this field, having different perspectives and domain expertise is crucial, making it essential to not ignore other areas of one's skillset. As a result, it's vital to be intentional about showcasing one's entire package of skills, rather than trying to sell only the new things we bring to the table.

**Finding Answers in Uncertainty**

When faced with uncertainty and questions that don't have clear answers, it can be tempting to feel overwhelmed or stuck. However, speaker Hillary offered a valuable perspective on this topic: "Let go of the idea that you have to have a definitive answer before you could proceed on with your life." Instead, she suggested taking a more flexible approach, such as researching online, following experts, and reading blog posts.

This mindset shift can help us navigate the uncertainty that surrounds us. By being comfortable with the unknown, we can begin to see opportunities and possibilities emerge, rather than getting stuck in a rut of indecision.

**A Time for Growth and Exploration**

As we move forward in this uncertain time, it's essential to remember that growth and exploration are key components of career development. This conversation highlighted the importance of embracing uncertainty as a catalyst for personal and professional growth.

Rather than trying to optimize every moment of our lives, we should focus on cultivating a mindset that is open to possibility and discovery. By doing so, we can create space for new experiences, learning opportunities, and connections that might not have arisen if we were feeling stuck or uncertain about the future.

**A Message from the Panel**

As our conversation came to a close, one speaker shared their thoughts on how individuals can find answers to their questions: "Google it, follow people who are good on Twitter, do some research, and if it turns out that no one has a consistent answer, there's probably no answer." This approach emphasizes the importance of seeking out diverse perspectives and resources, rather than relying solely on one source or expert.

In conclusion, navigating uncertainty in our careers requires a mindset shift. By embracing the unknown and being open to possibility, we can create space for growth, exploration, and connection. Remember, it's okay not to have all the answers – sometimes the most valuable thing is simply taking that first step into the unknown.

"WEBVTTKind: captionsLanguage: enwelcome to the twiml ai podcast hey everyone in my message on monday discussing george floyd and the fight against racism i mentioned the responsibility we have as machine learning and ai practitioners to ensure that the tools we're building are fair and responsible and don't reinforce racial and socioeconomic biases as a follow-up we've curated a playlist on the topic of fairness ethics and bias and machine learning and ai topics we discuss frequently here on the podcast i encourage you to check these episodes out and engage in conversations about these issues with your teams and in your work in addition on monday june 8th we're hosting another interactive viewing session this time i'll be joined by my friend rumon chaudhry managing director and global lead of responsible ai at accenture please join us to watch the interview and for a live q and a session in the chat to register head over to twimlai.com 381 viewing and now enjoy today's show which comes from the recent panel discussion we held on advancing your data science career during the global pandemic hey everyone welcome to our program today i am really excited to be joined by an amazing panel to take on the topic of advancing your data science career during the pandemic my panelists today are caroline chavier caroline is ceo of the alliance.com and co-founder of paris women in machine learning and data science anna maria echeverri anna is a ai skills growth and strategy lead at ibm hillary mason who is working on her next venture and is formerly the gm of machine learning at cloudera fast forward labs and jacqueline nolis jacqueline is principal data scientist at brightloom and author of build a career in data science so i don't think we need a lot of context setting for this discussion we've all probably heard the word unprecedented thrown around a lot more than we can stand but that doesn't make it any less the case as of early may here in the u.s at least unemployment had risen to almost 15 percent the worst that we've seen since the great depression with over 20 million jobs lost in april but beyond the magnitude of these numbers the time frame in which this is all played out has made things extremely challenging uh we've seen changes that might have otherwise taken months to play out during a quote-unquote regular recession if that means anything happened in just a few weeks uh starting in uh in march certainly many in the data science community have been impacted by this situation we've seen senior folks data scientists and machine learning engineers impacted by layoffs at communities that just a few weeks ago couldn't hire enough of them we've seen new graduates finishing their degrees and boot camps looking for opportunities during very challenging times and we've seen folks with secured offers and internships who now need to reevaluate their plans so today we're here to talk about what you can do as a data scientist to navigate this uncharted territory from a career perspective before we dive in a couple of quick notes uh first i want to send a huge thanks to our friends at ibm for sponsoring this discussion ibm is committed to educating and supporting data scientists and bringing them together to overcome technical societal and career challenges through the ibm data science community site which has over ten thousand members they provide a place for data scientists to collaborate share knowledge and support one another this is a great place to connect with other data scientists and to find information and resources to support your career go to twimlai.com ibm community to join and get a free month of select ibm programs on coursera next we want to make sure that this is an interactive discussion we'll be keeping an eye on the comments that you submit via the youtube chat and it's really my sincere hope that you drive a good part of our discussion today finally we will be bringing you more discussions like this on a wide range of topics so if you'd like to be notified when we schedule future conversations head over to twimlai.com newsletter and subscribe there so i'd like to get started by having our panelists introduce themselves why don't we start with you caroline tell us a little bit about your background and the perspective that you're bringing to our conversation today yes thank you sam um so my name is caroline xavier i am a french tech ai recruiter i have been one for the last seven years so tonight i hope to share very pragmatic pieces of advice and you know have another perspective as a recruiter i am not just a recruiter i also co-founded the paris mlds chapter vmlds stands for women in machine learning and data science so it's a global organization which aims at promoting and supporting women and gender minorities in machine learning and data science so it takes a lot of my time but recently i also founded my own company called the alliance and via my company i want to help companies reach excellence in hiring what do i mean by that designing the best recruitment process engaging the best candidates having the most amazing diversity inclusion program and i know it's a tough um goal but i want to to go for it and i'm happy to be with you tonight awesome awesome thank you anna yes thank you sam so my name is anna chaveri and i work for ibm i lead our ai education programs team and i went into the world of data science about 10 years ago when i decided to go back to school and get a master of science and analytics prior to that my undergrad was a computer engineering degree and i had an mba and i had had multiple different roles in the technology industry services uh sales marketing and i decided i wanted to reorient my career into data science so that's what i did i'm excited to be here today in my current role i work a lot with internal and external people that are in the process of upskilling for for the world of ai which is going to be pervasive pretty soon um so i'm hoping to share those experiences i've had as i've been able to have these conversations with others and share that expertise here all right great and hillary hi everyone i'm hilary mason i'm pretty excited to be joining this discussion today i've been working in machine learning and data science for about 20 years starting first in academia and then transitioning to industry i like to build things starting about 10 years ago i've been primarily in management roles um leading data science teams as a chief scientist as a ceo of my own company as a general manager uh and then i started a new company in the last uh couple months which has been its own interesting experience um but one of the the great joys of uh having been around and being a manager is that i get to help launch some really brilliant careers and uh in this context that means i've been having a lot of really deep uh career conversations with folks across the data spectrum so pretty excited for this conversation today awesome and jacqueline hi uh my name is jacqueline nolas i have a background in mathematics i got undergrad master's in math phd in industrial engineering i've been doing data science for about 15 years now and probably about eight or ten of those as a consultant so working as a data science consultant becoming a director at a data science director at a consulting firm leading a team of data scientists realizing that i didn't like the management tax as much and much more like actually writing my own code and so then working as a freelance consultant for a couple years now but um with the uh virus making consulting independent consulting very difficult i uh last week started a new job at a company called brightloom as a principal data scientist so not only have i uh am i on a panel about advanced degree and the pandemic that i also have um and uh i also uh i recently in last year wrote a book with emily robinson uh build a career in data science all about how to become a data scientist or become a more senior one decide to be a manager um make all those sorts of career decisions awesome great so uh if you're in the audience you can see we've got a bunch of different perspectives here for uh to uh share with us so uh don't hesitate to get your comments in uh via the chat you know i'd like to just start with a a pretty broad question about you know what are each of you seeing and what is kind of different and unique about the environment that folks are facing right now from you know the the particular perspectives that you're coming at things we can start with you carolyn i'm not going to use the word unprecedented but it's tempting in europe as much as i have noticed um there was a decrease a significant one of the job position being published online meaning on linkedin and different job boards so it was a 60 decrease for instance in france um so there are two ways of analyzing it yes there was less than usual but there was still 40 of job offers being out there so i would say that was different for from a candidate perspective is that the the the number of opportunities was significantly less so i could notice that candidates were a little bit more creative in the way they communicated online and the way they gave visibility to their work so for instance i saw people getting back to the recruiters that wrote to them in the past and that they ignored and so it's interesting to see a switch from candidates not answering to recruiters but now they go back um and then i think i would like other panelists to to speak about that because i also got the opportunity to help data scientists looking for a job so i can also share with you the tips i gave to them awesome and how about you what are you seeing i think some of the main challenges that i'm seeing is the fact that a lot of uh growth plans have been put on hold so you know while you know we work with a lot of clients around the world and it's organizations that had significant plans to grow this year and hire additional data scientists to their teams are not necessarily saying this is not going to happen but maybe not right now so so i think that has a significant impact right because it's been reduced as caroline was saying and has been reduced the number of jobs out there have been reduced significantly i do think however that this is a temporary effect so so everything that brought the data science careers to the forefront and made it such a high demand job those things still exist out there i think it's just everyone is going through this adjustment process so uh if anything for you know from the perspective of people out there looking for new jobs it just means that they need to do more of what we're doing and and really focus on doing it better and and more being able to sell themselves better because the opportunities are going to be reduced at least you know for for for a period of time that we don't know how long that's going to be jacqueline you mentioned in our pre-chat that a lot of the process of finding your new position was the same but it was just much harder yeah how much harder elaborate on you know what you experienced so um years ago when i was looking for a job i actually tweeted you know hi i'm a data scientist looking in seattle area and i you know i got a certain number of like views of that tweet i got a certain number of leads and i tried doing that again about a month ago a month and a half ago and i got five times as many views and maybe a fifth the leads so like basically a couple years ago when i did it it's like oh what about this company this gun and then this time i was like have you tried amazon like it was just nothing and i was just astounded because my my resume had saved largely the same if anything it presumably got better um but it was just like it was just total silence um and you know i was able to get a job and i'm very happy with this new job very excited about it but just the number of opportunities is just minuscule compared to what it was a few years ago how about you hillary you're talking to a bunch of folks out there what kind of experiences are they having yeah i mean maybe i'll take the perspective here of people who are hiring um because i've spent a fair bit of time focusing on building teams i have a lot of friends and colleagues who are in that position and i think it speaks to the of course there are fewer positions open but the types of positions that will remain open will not be the ones where it's like let's hire some data people and throw them at the wall and see what they managed to make for us the ones that are open are the people who have a business generally at some scale where there is something to optimize and they know they need a certain set of skills to do that and those folks are gleeful about some of the talent coming on the market um so it's not you know all doom and gloom i'll just say from where i'm sitting i'm seeing from that perspective the landscape of opportunity that exists has changed pretty dramatically in the direction and there have always been a bunch of flavors of data science jobs some very research oriented some very you know you optimize one metric and that is all you do um many in between uh many you know in the the sort of you're the bridge from the the data and tech into a business unit like maybe you're working with financial um reporting um you'll see more more jobs that are a clear need where there's you know revenue and growth already attached and fewer of the highly speculative fewer research type positions right now and that's from the the folks who are hiring and creating those positions um so anna brought up an interesting point that uh i'll segue into one of the questions that we've gotten uh and that is that we're you know immediately prior to the pandemic was a period where you know data scientist was an extreme growth you know area growth opportunity growth career and as a result of that we've had lots of people joining training programs boot camps over the past few years traditional educational institutions have spun up data science programs so we've got not just folks that are you know experienced and and in positions impacted by layoffs but new data scientists and so we've got a question uh from willie costello you know what's your advice for new data scientists that are trying to break into uh the field for the first time anna so i talked to a lot of people in that same situation right they they've done the work of going back to school joining a book camp taking different online courses and really focused on on getting the skills to go into a job of data science my main advice would be do not stay there so one thing that i see that's very common is people that come to me with i've got this certificate or this this course my next question is always going to be how have you used those skills you know have you been able to bring them into your current job have you been able to come up with um topics or interests you and you know what have you done with the skills so my main advice will be as you work on acquiring the actual skills sort of from the learning perspective find ways to put those into practice because i think especially in a in an environment right now where the opportunities are limited we will prioritize people that kind of bring that element of not and it's not just the the sort of official formal real world projects but also the drive they they i did not just you know took take these classes but i just went in and came up with this interesting hypothesis and i built something around it right like all these like cool research innovative data science jobs they're really um they're the ones getting hit the hardest right the jobs that are like the jobs that a lot of them are staying around are the analyst jobs right and i think as a data scientist or if you want to get into a science field you it's often you lost some serious attitude like i'll accept nothing less than the title of data scientist but there are jobs that are not quite title data science but use a lot of the same techniques skills and are great building blocks to help you future your future grow your career and so especially the time like this where things are so volatile um finding a job that like meets the criteria of a lot of the things of data science uses numbers you have to play with num you get to play with data things like that but maybe isn't exactly what you want to do like that is so worthwhile um in a way that wasn't true six months ago on the the topic of building a portfolio and i'm going to come back to your uh your point in a minute jacqueline but on the the returning to the topic of building a portfolio uh myrtle janeau asks what kind of projects would you expect to see or want to see in a portfolio anna one of the things before i talk about the what type of projects i think it's very important to have your your github repository very much up to date so that's one thing that we usually look at and and i know you know when we're working on a resume we're more focused on what we put on the resume but a lot of times if you're not adding your github account this automatic systems can actually even discount your resume altogether before before they even bring you to a recruiter in our case at ibm because we everything that we do is so open source driven so i personally look for projects uh leveraging python because that's where a lot of our clients are working with technology so so i want to see people that have been able to uh really work across the whole you know cycle of building a hypothesis and preparing data and and building some models and isolating your best model using python but i there's something that i always look for that that goes beyond that technical capabilities and it's i like to be able to infer your curiosity levels and that's something that for me is very important because i find that people with very high curiosity levels tend to be really solid data scientists and and really are able to deliver a lot of business value so so it's not so much the you know i use this technique or i use this other technique i i like to see the process i like to see how you thought through uh a question that you had a hypothesis that you had and how you uh brought that to something that would allow someone to bake to make a better decision around that topic i can jump in with one observation as someone who's done quite a lot of hiring requiring people to put things on github is really difficult because certain people are open to harassment if they do that just due to the nature of who they are and so i'd really encourage folks to take a step back and say what do we want to get out of seeing a candidate that has written code in public um and are we going to provide other ways for them to demonstrate that they are capable of writing code uh that they accomplish things that they have their name on things um and so i'd really encourage folks especially people hiring to be very thoughtful about what you require on those applications because you will be excluding a fair number of people from your candidate pool or forcing them to take on some kind of risk that really isn't appropriate or evenly distributed but then i'd also like to reinforce the point about creativity and curiosity and i think that in so many discussions of data science we focus on the technical capabilities like do you know pytorch you know python um but actually what makes a great professional is that curiosity and creativity and we assume you're going to be able to learn what you need to learn as you go and like frankly nothing that i use today existed when i was in school um and so really looking for folks who have the ability to know when they should start to learn something new too awesome carolyn did you have something to add i agree with both anna and hilary um my advice is always to harmonize everything that is going to present your background and your work meaning i don't want to find different information whether it's on linkedin or github or kaggle so first no matter if you're a junior person jumping into data science or senior person try to look for harmony in what you describe about you and i agree with what hillary said because there are biases and we should not expect all candidates to have public repo repositories however i always encourage candidates to try to build the bridges between the expectation of a company you know look at their blog post see what kind of technical issues they face the technologies they use and even if you have a different kind of background try to narrow the the difference and make it easier for a recruiter to point out why you are a relevant candidate it can be curiosity it can be the time you spend on training yourself it can be you know i was teaching kids how to code this stuff for this kind of stuff we all need to value what we do and just to build a little bit further on all these really great points i mean i think it's just like just be passionate about it like one cool blog post or like one i saw like one time i hired a person because she made a really cool power bi dashboard about a theme park and it's just like i love that stuff right like just one small interesting thing that's going to hook me in does so much better than a jupiter notebook with 10 000 likes of python showing that you use tensorflow i guess but i don't know because i can't really follow what's going on in it like just if you are gonna put something out there just make it like make it something someone's you know gonna be a little intrigued by um and like that involves like really just like being curious as was discussed but like yeah just like add some zest like make it about you jacqueline did you find that the portfolio what played as a bigger role in your recent job search as a more senior candidate um you know it was fun in interviews to be like oh actually i wrote a book um that was fun um the thing that so here's the interesting thing i don't think very often that like necessarily they're like oh i looked at your github page and i saw this package and you know i looked at line 27 and i saw you really did something clever there so i'm gonna hire you it wasn't like that but what was happened was like they loosely saw oh you'd have some cursing like the cursory size and knowledge on this like i like for one of my side projects one time i uh made a neural network that i trained to generate offensive license plates um using a historic data set of arizona offensive license plate and it was like i never used a neural network before and this taught me neural nets and then i went on to implement that at like a massive company in like a huge ml pipeline and i only learned it because of that silly side project and so what did the company do with all those offensive licenses that you generated like just this knowledge you know just learning this stuff really helps and them seeing that like oh you clearly like i gave a talk about this and that caused that you know like like just seeing the excitement build really goes a lot farther than showing that you know a particular technique or a particular couple lines of clever code so we've talked uh about things for people new to data science we've talked about folks that have experience carolyn have you run into folks that are career switching so they've got experience in some career but are trying to transition into data science and do you have any unique perspective for them yes indeed and on this topic i have a lot of uh beautiful stories thanks to the umlds chapter in paris because um for instance we chose to run our meetups only in english to be inclusive to foreigners joining paris and we have helped astrophysicists mathematicians wanting to switch from their initial field into data science so i would really encourage people to connect with the support group it can be our ladies we mlds women who code um even the online platforms such as twitter kaggle sometimes stack overflow can be useful on this topic as well but you know to connect to people who might have done the same switch as you to first know that people have done it so you can do it to get also insights of where are these people working so it means that possibly these companies could be hiring you and i really believe in the strength of network offline and online and right now with the pandemic you have to make it as much as much as possible online so these support groups are existing so if you want one example because you know i don't want to just give ideas i want to give working ideas we had a mathematician she was american and she came to paris and you know she didn't know where to send her application so one of our members referred her in her company and at the end of the day she got hired and the thing is she asked for help in order to prepare herself for the phone interview with the recruiter because it was the first time for her that she was going to present her background for a new job that she hasn't done so this is something ask for support and don't try to remain isolated one of the topics that came up in our our pre-chat was the idea that you know there's a difference between a data science job and a good data science job and that uh as anxious as we all are about you know unprecedented times and um you know as important as as jobs are that you know there's uh if you can be a little bit more discriminating in the way you pursue opportunities that will have long-term benefits for you i think it was you hillary who originally brought that point up can you elaborate on it absolutely um one of the things that i find somewhat astounding still about data science is that if you go read data scientist job descriptions at a bunch of different companies they're all different um and you'll realize like once you start looking at these positions that they actually uh work really differently so some are centralized teams sometimes you're the only data scientist on your team sometimes you're in a group that's responsible for operational analytics and making some predictions but you're not doing machine learning sometimes you're in a research team uh you sit in different places in the org you have different types of managers and management experiences um and then you layer on top of that all of the complexity of choosing a company to work at any way so you know is it a supportive culture a competitive culture a political culture um do they really invest in their own people and grow their careers or is it more of a burn them out churn them out sort of place um so there are a lot of things to consider when you think about what's a data science job and what's a good data science job and what's the best data science job for you um and it is always hard to find those good and best jobs under any circumstances much less now and so i'd really encourage folks to be very thoughtful and ask a lot of questions about you know how does this team work together am i going to be the only data scientist on my team and by the way if you're entering your career try to find a place where you can work with a team of people who have been doing it for a while because your first job is going to set you up for your second job and your third job and the more you learn from people with experience the faster you'll be able to grow not as much a concern if you're very senior and you're quite happy being the only data scientist on the team um so it's really they're all of these things to be thoughtful about even beyond what is my title what is my salary you know sort of how do we work together and i think we don't in our community we don't really talk about those things nearly enough can and on that point um it's gonna take you a while before you figure out what are the things you want in the job like i think the first seven years of my career i was switching jobs because like no i don't like this part no i don't like this part no i don't like this part like it really took me a long time before i really got comfortable with like okay here are the things i have to have in a job your things that are nice to have the things i don't care about and like for me now this bar in my career like the thing that i need more than anything is if i'm like hey this is a good idea or hey this is a bad idea i actually have a chance of influencing the like the trajectory and sometimes i got a big company you may not get that so that's something that's really important to me but i know lots of data scientists who like if they just get to be in their corner doing their data science they do not care if they get to influence the like long-term term trajectory of stuff so i think that point's totally right i think it's fine like if you get into a job and it turns out you don't like something about it that is not a failure of you that is a very normal natural part of this process and like it's just a it's a whole growing process and these things change too as you get more senior and your life goes on what your priorities are may change too i also like to think about it in terms of the things you hate less than other people hate them so like maybe you may be at a point in your life where you don't mind working 60 to 80 hours a week you may be at a point where you don't mind travel once we're allowed to do that again but you may be politics doesn't bother you a bit or that you're great at office politics um and so it's really figuring out yeah what flavor of company and culture is gonna be one where you can drive and and i would add to that hillary what flavor of a job in the data science ecosystem because this is this is not you know i think years ago let's say let's use another example you were a database administrator that was your job and it was very clear and i think when it comes to the field of data science machine learning artificial intelligence it's all this ecosystem of people that may have different roles at different stages throughout the build out of the solution so you know you have business analysts and you have developers that are maybe consuming the output of a model and integrated into an application you have people that are managing these models in production and so so i think a lot of one recommendation that i would give to people in the audience is explore all these sort of adjacent roles do do not necessarily go with a job title when you're applying for a job because as hillary said you you put 10 companies and 10 job descriptions for data scientists and i we can assure you that they're all going to be different so so do not go just by by just a job title break that down into what it entails what a job entails same with your skills instead of saying oh i'm a data scientist or i'm this or i'm that really break it down to the skills that you have and and bring those forward in a in a conversation and an interview because i think at the at the end the most important thing is what can you do what are you able to accomplish and and that uh the the name for that umbrella may be different all over the place could i have one quick thing which is to say also i think we also have a notion that you have to progress on a ladder like you have to be a junior data scientist and a data scientist and a senior data scientist and then maybe you face the existential crisis crisis of like management and that philosophical uh change in your role and at least i may be a bad role model but i've been a data scientist a software engineer a ceo a corporate executive the world's shitty assisted men like you can bounce back and forth uh and do this pretty successfully so look at your contribution yeah and i really don't think i think that's true and i also don't think that one is necessarily better than the other i think for some reason there's a stigma that like machine learning engineering is superior to decision science or data science is superior to um analysts and like there's a lot of these con preconceived notions that are not true and also like if you if you dig into them a little bit they're really messed up um so like don't let the idea that like oh i i am i'm not that job is beneath me because it's decision science or i can't do that job because i'm only a decision scientist like don't let that stuff hold you back caroline i've got to imagine that as you work with candidates you are faced with you know folks that aren't sure exactly what they want to do or folks that have lots of different job offers and need to choose between them is there a you know a framework that folks should you know keep in mind as they're trying to uh you know either figure out what they want to do or evaluate different opportunities or point themselves in the right direction career-wise good question first of all i think good candidates are informed and prepared candidates meaning i will second what hillary said earlier before starting to apply to anything have a look at the salary ranges the companies hiring in the field you're targeting i don't know if you have been enjoying psychic learn or numpy and you want to keep working on these libraries look at the kind of applications and companies using those look at the experts in the field and maybe try to connect with them because right now they have much more time to dedicate to peers aspiring data scientists to talk about their job and how you know they built their own careers and usually you know it's a recruiter technique um don't write it to someone saying hey i'm looking for a job you can do that if you are if you're okay but i think it's too aggressive but write to someone hey i am wondering if would you mind sparing me 15 minutes on a zoom or hangout call and try you know to understand what you would be enjoying on a daily basis so this applies to junior people and then if you are experienced the thing is um i'm always thinking about this uh article the engineer manager pendulum from charity myers you know the more you grow in the management ladder the more it's going to be narrow for the to look for your next step so while you know that get ready for describing your technical projects get ready to talk about your aspirations get ready to talk about the technical issues you faced yes i was modeling this algorithm it was kicking ass on the paper and then my my data was noisy so here's the solution i found prepare that as much as possible because no one can do that better than yourself especially if you have a recruiter in between you and the hiring managers and the future team members this might be a little bit less valid if you directly speak with a peer so that being said ask a colleague that you trust to make a fake interview with you to evaluate how much you need to get ready or you can list questions and ask for a member of your family to ask them to you and you hear yourself you correct yourself and you become ready when you know you want to do great as an interview so these are pragmatic pieces of advice i would like to to point because um the crisis stressed something we are going to prioritize the candidates and the hunt is even more difficult than before i want to take a few minutes to jump into i mean i've been following pulling questions in from from the chat throughout but i want to jump into a few of them a lot of the questions are of the um hey do i need an x you know and an x might be a phd or a kaggle rating or um a portfolio we've talked quite a bit about that you know any specific responses to you know phd or kaggle ratings or i'll tackle i'll talk about those so i have a phd phd not good idea don't get it i mean not if you want to go be a data scientist in industry um just because like oh my god it is so many years of your life it is so much lost income and the percentage of stuff you learn that's actually useful to working in industry is very small there is some value you learn how to learn you learn how to kind of like do some portfolio building when making a dissertation but nothing compared to actually the path of going into industry just before a phd which is not to say that no one should ever get a phd or the phd won't ever help you but just if your question is like i really want to go into data science i don't know if i need a phd or not like no no don't beware um and then the kaggle score um this is kind of this is like it's like the same thing with the github account where it's like it comes this like weird metric where like oh you need account you want to apply you don't even have a kaggle score right you don't even have a github and it's like no you don't need these things these things can be helpful right being able to say like look i did this kaggle competition i did quite well i can make a story around it that could be helpful in an interview just like a github portfolio can be helpful in like look here's some things i can show you how you can did it but there are a thousand ways you can show a person that you are capable of something and the idea that you have to use a preset thing right like you could right if you go to a conference you give a talk you can make a poster you can make a cool website just for you there's like a thousand things you can do that are show and interview you have more capability than just what shows on a resume and none of them have to fall into github kaggle doctorate um and don't don't can can i add one thing to that there's one thing that i actually find very useful and it's for someone to have their elevator pitch so if you only have one minute with me if you come up you know and send me an email on linkedin for example tell me why i you know i should be interested in talking to you tell me what you think is the value that you bring to the table and and that can be done in you know literally a few sentences and it's much harder to do than it sounds it's actually much harder to do than if i tell you talk to me for an hour about your life story because you really need to bring it down to the most important things but i but i do find that that's very valuable as you know of course now we're all like locked in our houses but if you're in a up and you can talk to someone you can tell them hey this is what i do this is the value that i bring to the table i find that that every person should have their elevator pitch ready uh when they have that opportunity via digital via a face-to-face meeting you know whatever that is and i know we sometimes don't spend enough time thinking about that because we spend more time thinking about the project and the technical skills and all that and it's like summarize to me your value in a few sentences and i can guarantee that that will get you you know doors open uh because it'll be easy to understand that's a nice way to just sort of put a framework on this question which is do i need an x is usually about opening the door there's how you get through the door which is how you get through an interview process and then there's how you actually are amazing at the job these are three completely different things correct um and any of these x things can help you open a door but there are a lot of ways to open a door that don't involve having a kaggle score um don't involve necessarily having a certain degree of any kind um in fact with kaggle stuff i personally um and way more interested in the questions someone is asking themselves so it's cool if you've done a bunch of kaggle competitions but they give you the question i want to see you know maybe uh you know i did this kaggle competition and then i got really curious about this question and then i found this other data set and i took it off in this direction um so it is uh you know whatever you find to open the door um and that could be a good elevator pitch it could be a twitter thread it could be a visualization you make um there are lots of ways to open doors and there's no one right way and there are certainly some that cost a lot more in terms of time and effort than others and i think when it comes to making an elevator pitch like i remember when i was first finishing like first going to get my first job thinking like like i don't know how to make an elevator pitch like there's nothing special about me i got a degree and i want a job doing this stuff like what is there like how do i say and i think this is where that portfolio goes a long way because by like reading doing side projects reading they're checking twitter like there's like you start to kind of understand like what's out there what is more interesting in you and just being like look i'm a data scientist i just finished this bootcamp i'm really you know i'm fascinated by time series analysis um and i'm looking for a job like just having anything like like one like like like kernel of something interesting about you is just a huge difference yes yeah i'll add to the one of the ideas that i found really helpful uh in thinking about careers is something that i picked up from scott adams the creator of the dilbert cartoon he talks about this idea of talent stacking which is essentially you know if you want to be successful you can take the route of hey i'm going to be in the top one percent of whatever field you're going after or you can be pretty good at a couple of things and so his example is you know he's not a great cartoonist he's not a great writer but he's pretty good at both and he mixed them together and he created this amazing thing uh dilbert and there are tons of you know skills and talents out there that you can remix to create a unique um you know unique positioning for yourself caroline you had something to add yeah i am a massive twitter lover and you know sometimes even if you don't know how to talk about yourself you can highlight other people and get visibility about your work and i completely agree with jacqueline like there's plenty of ways of being creative and talking about your work when people write blog posts or now you know the younger data scientist they do tutorials on youtube i think it's great because you know it's not just you show off about yourself you you share knowledge with other and this is how um i think you can get attention so please be vocal on twitter on discord on the different slack accounts you can join and don't forget to give visibility to your work and yourself and if you don't want to promote your own work make a network of people who all can promote each other's work because there is nothing more fun than promoting your friend's work any of you have recommendations for folks that want to take this opportunity to kind of skill up and you know better position themselves for their next thing is there a genetic why could i do could i do just the opposite right so like i was a consultant i was working crazy hours uh the consulting contracts went down and i was waiting you know to my next job that absolutely could have been a time where i learned all these new skills like i'm even better but i could tell that in fact i've been on the verge of burnout for years and so i didn't do that and i used this trying time to make sure that i just stayed mentally and emotionally healthy and really trying to handle that rather than trying to like see if there's just one more skill i could get and i think that ended up being really helpful for me in a way like taking hour long walks was probably more helpful than an hour of learning new tensorflow stuff um which is not to say that leveling up your skills and like that stuff's not great like definitely if you want to do that that's good but also the idea that you have to use the quarantine to do this um i think can be really difficult um and maybe don't do that yeah i i agree with um jack let's not all think that we should be uh productive but my mom used to say um try to transform the negative into something positive so personally you know i was a bit depressed of not going to the gym not being able to go out and so i was like what are the things that i usually push back because i have something more important to do and i did these things and when you are a data scientist it can be updating your different platform it can be reaching out to people it can be joining a platform because you thought you were not good it can be learning how to put code on github because i know that women for instance are underrepresented on github for the reason we mentioned earlier so i like to try to transform something bad into something that you can be proud about later and it can be a work outside well and i think to that point like a lot of things are not data science skills but still useful like i updated the html on my website during the quarantine and like that's like learning more html isn't like data science skills but like that actually helps me when i eventually make a dashboard things like that so there's a broad set of things that could be useful anna how about you you're a lot of your role focuses on skills development any takes on this one yeah so so for me the uh so i get approached a lot by people both internally at ibm and externally that tell me would you spend some time with me i want to become a data scientist can you give me some guidance and and when i do that i kind of flip the conversations to not just telling them this is how you become a data scientist but tell me about you tell me why you're interested in becoming a data scientist so this is an advice that i would give to people in the audience too which is you know break it down into the things that you love and a lot of times i find you know i would say more often than not as we start kind of breaking apart kind of the skills that you need to become a data scientist do you know machine learning i find that a lot of them are really not into that they just felt attracted to a piece of it so i'll give you an example some people tell me oof i could spend my day cleaning data i love to clean data and i love to you know just bring data sets together and so that's more the skill of let's say a data engineer persona and i won't call it a title i'll call it like a persona so if that's your passion you don't need to go and learn all day you know go back and refresh all your math and and really learn machine learning because that's not what you're interested in in other cases i find that people are just interested in developing really cool applications so they're a developer now and they feel like maybe i need to become a data scientist in order to build these really cool applications well maybe not you know maybe if you're not building a machine learning model from scratch if you're really your end goal is to build some cool applications using maybe some pre-built models maybe some pre-built technology focus on that so so i would my advice to people in the audience would be try to break things apart to step away from the data science as a title and kind of bring it down to what are the elements of data science that i'm interested in and maybe they're shortcuts to it to where you really want to be that that are not necessarily uh becoming a data scientist joseph asks um he's run into widespread bias towards older applicants but certainly there are other groups that are biased against in the the job process you know any comment or advice around overcoming those kinds of biases that's a really hard hard drive someone that's a female hispanic and on the older side like i have three strikes right so it is it is challenging and i i personally don't think there's like a magic wand to get through it um i personally focus on my skills the things that i can do the things that i'm really good at uh but that's easier saying that said than done unfortunately if you encounter a situation where you feel like there is biased against you for whatever reason don't waste time on it and most importantly don't let it don't take it personally so just you know there are many jobs out there you only need one of them just keep going and also share information about that employer with other people in your network so they don't waste their time um but you can't fix someone else's bias certainly not as a candidate how about you caroline yes um hillary for the win i want to say um i would add that if you are coming with an internet intersectional background meaning you face different discrimination because you're a black woman transgender um and you face this issue i agree with them i really don't go however there are like groups existing that can help you again to select the right places for instance there is queer and ai blackened ai latin in ai lgbt in tech for instance so i strongly encourage people to go and look for support um because you know um i am a lesbian myself and sometimes you're like no i'm totally imagining something and then you discuss with someone facing similar issues and you realize how much it's not just an emotion it's a fact and um if you don't want to work with this kind of people so other people made a work prior to you and go and see who supports the which companies support these communities and who are the people in these communities that can refer you or talk with you to reassure you refer you for the the people who are more privileged listening to this right now um so i started on the my career path as a white man and now i meet trans women and so the be like one of the biggest things now because this is the first time i've been on the job market as a trans woman and the thing that's so different is now when i'm at a job interview i have to spend so much effort being like are these people gonna be good to me when i'm like when i'm here like how many women are there like i'm doing all this mental math that i just did not have to do as a guy i just like was like oh cool this job seems nice i like the tech i'm here and that burden is real it's on like minorities all the time and it is the worst um so it's just like this is like i totally the person asked the question what you're dealing with is totally valid everyone who's in these situations is totally valid if you're like oh that's not real um because i haven't had to deal with it that probably means you have not been on the receiving end of it but it is definitely valid and real yeah i was just going to say that in this field diversity is so important so so one of the things that i that i would encourage you as a job seeker is to research the companies that you want to work for because there's definitely a difference i mean some companies are really more focused and more intentional about building diversity in general but especially in data science teams because we don't want a team of the same people you know uh kind of driving decision making in the organization or influence influencing significantly decision-making in the organization so very important to research the companies because you will find the difference in in terms of their diversity and inclusivity efforts i want to make sure we get to a question uh by juicer having done a bachelor in statistics data science seems best for him but a lot of universities are adding a master's in cs prerequisite uh so what's actually the most important aspect of positioning yourself we've kind of come at this from different directions is there an ideal set of skills or benefits so i will say this um i started my career long before data science even existed as a term i think most of us here are judging by everyone's years on the field and so for companies for years had to grapple with the fact that there's no set list requirement for what is a data scientist and that companies we're just starting to get to the point where you can get a degree in data science but it's nowhere near normalized like that is not standard at all and like some company you know some schools yeah maybe they require cs degree undergrad maybe they don't but if they are putting these strict requirements in there they're making big assumptions that are probably not true like the fact that you need a cs undergrad to do masters then data science is like requests um and so and if similarly if a company is like oh we can't hire you you only have a stats degree to do data science that's mind-boggling so just you know just like if the companies are doing this that is more of a sign that they do not know what they actually want on a team and they're probably not a team you'd work want to work for then it is a sign of like oh you really did not take the right four year path or the 15 years of tensorflow experience yeah come on what about the the process during kind of shelter in place we've talked a little bit about how networking is harder interviews are different now any tips for managing that caroline what are you seeing in terms of the way the process has shifted uh today yeah so massively you know companies that were usually reluctant to have skype interviews or video oriented interviews are now completely open surprisingly and this is also adding another layer of difficulty for candidates because you know when you interact with someone the body language counts a lot of the percentage of how convincing you're going to look and sound so it might sound super pragmatic but while going through offline interviews make sure to look at the camera make sure to have a working microphone make sure to have something to take notes in order to keep up with the flow of the conversation these might sound silly but you know i've had like people you know like that and then you have the gardener coming with the in the back in the background this is not possible this is not professional if it's a kid it's okay but anyway this is different and i wanted also to suggest something that we mentioned with meetups luckily meetup groups are going also online so if you want i would invite people to reach out to meetups organizer to try possibly to introduce their work because it might be a little bit less impressive to present it while you're at home comfortably seated instead of having like uh hundreds of faces in front of you so this is another negative thing that you can switch into an opportunity and you thoughts on taking the process virtual i would just say to that point um the lack of conferences is really hard but like twitter data science twitter is so good like i really get a lot of value meeting people replying to them and it's just just twitter that's the social network for me um so i'd highly recommend if you're a data scientist you want a network you can't go to conferences and you're not on twitter then like definitely get on twitter awesome awesome so we're uh winding now now as the last few questions come in um maybe i'll have each of you take a moment to kind of your top three words of advice or top couple of you know takeaways the number one thing you would do whichever way you'd like to go with it uh hillary great um i think my top takeaway or a bit of advice is that um at least a lot of the folks i've been talking to now and my friends who are job hunting are doing so in an environment that has a ton of anxiety uh in it and of course making major life decisions while you're suffering through anxiety and uh you know stuck at home dealing with responsibilities there is always a terrible idea so the one bit of advice i'd say is to try to keep things as systematic and objective as you can to remember that you are awesome there are a lot of people out there who will help celebrate you sam being the one who kicked all of this off by retweeting people's accomplishments i'm happy to do the same um and uh you need to set down some criteria and rules for yourself about you know what you need to make on top of that what are the things that you prioritize in a position that'll make it one where you can actually really succeed over time um and then systematically approach finding it and trying to do so in a way that your own anxiety doesn't drive you to weaker choices than you might have otherwise made and i know that's very hard to do so rely on your friends to help you there too carolyn yeah on my side i would say book discussions with peers read watch tutorials make something that is going to inspire you and i'm going to quote jacqueline walk outside if it's your thing and i wanted also to stress successful stories um i have helped a student an indian student who got you know uh abandoned he was supposed to do an internship in amsterdam and he reached out to me and i just connected him with my indian network and he just uh found um an internship another female member of umlds reached out because she was looking for a phd she did a linkedin post a twitter post and she got hundreds of answers so don't give up just try stuff and if it doesn't work it's okay it's like a massive a b test that we are all going through and we are going to learn the way echelon yeah so i think what i'm going to say is like kind of a rewording what hillary said like i really believe it um you know so i just got off the job market because i was looking for a new job and i really like i had to make a lot of like like letting go ego things so i'm very happy with the job i got principal data scientist like small company like really was looking for but like i had to like really like i was checking linkedin and i'd like say like i guess i'm ready to take a job at like a lower title that i'm used to and like i have to give up like that's a lot of career progression i was giving up when i was making those decisions and it's just terrible and there's not like a cool thing i could say on a panel that makes that easy and it's not like there's anything about me that made it happen it was the environment it was the virus but that's the reality of the situation and i think the more you can dissociate that from your own well like sense of worth like the more i can be like i'm still going to be good to see a scientist even i'm still going to love it even if i'm a senior data scientist instead of a principle things like that the easier it is but that's not something that's easy to do and it takes a lot of practice and time and like you shouldn't beat yourself up over it if you are having these sorts of struggles anna um what i would say is that this is a multi-disciplinary field so as you are out there let's say you're transitioning from one career to a data science career or just embarking on it make sure that you really sell all the other expertise that you have as it relates to data design so i'll just use an example you worked in marketing for 10 years and you've been upskilling yourself and you you know you came up with a couple of projects in your company but you've only been doing data science for a year um i would say your value is it's exponentially higher when you bring together all the marketing expertise and your data science knowledge so a lot of times in our efforts to switch careers we want to oversell the new thing and say i've been doing all this and kind of ignored the rest i would say in this field at least from my perspective because it's such a multidisciplinary field because having different perspectives different domain expertise is so critical make sure that you are not underselling your whole package of skills one meta question we've gotten we've we've had a bunch of questions come in i tried to identify the ones that have the most broad impact but a lot of people have very specific questions i'm in situation x i'm thinking about why what do i do any thoughts on how uh folks can get those kinds of questions answered sure i'll take a stab google it follow people who are good on twitter do some research and if it turns out that no one has a consistent answer there's not a quick answer there's probably no answer right like the question i was just saying should i drop down to a senior data scientist because of the virus like how will that impact my community no one's gonna know that yeah like i don't know that like are you gonna know that sam like no one's gonna know that and i think the idea that each one of these questions needs an answer before you could proceed on with your life is like that like let go of the idea that you have to have a definitive answer do some googling follow experts um read blog posts but don't like don't try and like optimize every little moment of your life yeah i think that echoes a point that hillary made in our uh our pre-chat um you know this is a time of tremendous uncertainty and we don't you know we don't know what's coming we've not experienced this before uh and so to a large degree kind of pursuing a career moves in this time is going to require being at least uh being comfortable with that certainty uncertainty at least to uh some degree oh i also should say oh and buy my book that'll get you the answers that's the one way that's it awesome well uh i'd like to thank all of you for uh participating in this panel what a wonderful discussion and thanks everyone for contributing your questions in in the chat once again thanks to ibm for sponsoring this discussion we will have the video well the video will be up on youtube as soon as we're done at the same url so be sure to share it with your friends and you know feel free to continue the conversation in the comments thanks so much for participating everyone thank you all right everyone that's our show for today for more information on today's show visit twimlayi.com shows as always thanks so much for listening and catch you next time wow youwelcome to the twiml ai podcast hey everyone in my message on monday discussing george floyd and the fight against racism i mentioned the responsibility we have as machine learning and ai practitioners to ensure that the tools we're building are fair and responsible and don't reinforce racial and socioeconomic biases as a follow-up we've curated a playlist on the topic of fairness ethics and bias and machine learning and ai topics we discuss frequently here on the podcast i encourage you to check these episodes out and engage in conversations about these issues with your teams and in your work in addition on monday june 8th we're hosting another interactive viewing session this time i'll be joined by my friend rumon chaudhry managing director and global lead of responsible ai at accenture please join us to watch the interview and for a live q and a session in the chat to register head over to twimlai.com 381 viewing and now enjoy today's show which comes from the recent panel discussion we held on advancing your data science career during the global pandemic hey everyone welcome to our program today i am really excited to be joined by an amazing panel to take on the topic of advancing your data science career during the pandemic my panelists today are caroline chavier caroline is ceo of the alliance.com and co-founder of paris women in machine learning and data science anna maria echeverri anna is a ai skills growth and strategy lead at ibm hillary mason who is working on her next venture and is formerly the gm of machine learning at cloudera fast forward labs and jacqueline nolis jacqueline is principal data scientist at brightloom and author of build a career in data science so i don't think we need a lot of context setting for this discussion we've all probably heard the word unprecedented thrown around a lot more than we can stand but that doesn't make it any less the case as of early may here in the u.s at least unemployment had risen to almost 15 percent the worst that we've seen since the great depression with over 20 million jobs lost in april but beyond the magnitude of these numbers the time frame in which this is all played out has made things extremely challenging uh we've seen changes that might have otherwise taken months to play out during a quote-unquote regular recession if that means anything happened in just a few weeks uh starting in uh in march certainly many in the data science community have been impacted by this situation we've seen senior folks data scientists and machine learning engineers impacted by layoffs at communities that just a few weeks ago couldn't hire enough of them we've seen new graduates finishing their degrees and boot camps looking for opportunities during very challenging times and we've seen folks with secured offers and internships who now need to reevaluate their plans so today we're here to talk about what you can do as a data scientist to navigate this uncharted territory from a career perspective before we dive in a couple of quick notes uh first i want to send a huge thanks to our friends at ibm for sponsoring this discussion ibm is committed to educating and supporting data scientists and bringing them together to overcome technical societal and career challenges through the ibm data science community site which has over ten thousand members they provide a place for data scientists to collaborate share knowledge and support one another this is a great place to connect with other data scientists and to find information and resources to support your career go to twimlai.com ibm community to join and get a free month of select ibm programs on coursera next we want to make sure that this is an interactive discussion we'll be keeping an eye on the comments that you submit via the youtube chat and it's really my sincere hope that you drive a good part of our discussion today finally we will be bringing you more discussions like this on a wide range of topics so if you'd like to be notified when we schedule future conversations head over to twimlai.com newsletter and subscribe there so i'd like to get started by having our panelists introduce themselves why don't we start with you caroline tell us a little bit about your background and the perspective that you're bringing to our conversation today yes thank you sam um so my name is caroline xavier i am a french tech ai recruiter i have been one for the last seven years so tonight i hope to share very pragmatic pieces of advice and you know have another perspective as a recruiter i am not just a recruiter i also co-founded the paris mlds chapter vmlds stands for women in machine learning and data science so it's a global organization which aims at promoting and supporting women and gender minorities in machine learning and data science so it takes a lot of my time but recently i also founded my own company called the alliance and via my company i want to help companies reach excellence in hiring what do i mean by that designing the best recruitment process engaging the best candidates having the most amazing diversity inclusion program and i know it's a tough um goal but i want to to go for it and i'm happy to be with you tonight awesome awesome thank you anna yes thank you sam so my name is anna chaveri and i work for ibm i lead our ai education programs team and i went into the world of data science about 10 years ago when i decided to go back to school and get a master of science and analytics prior to that my undergrad was a computer engineering degree and i had an mba and i had had multiple different roles in the technology industry services uh sales marketing and i decided i wanted to reorient my career into data science so that's what i did i'm excited to be here today in my current role i work a lot with internal and external people that are in the process of upskilling for for the world of ai which is going to be pervasive pretty soon um so i'm hoping to share those experiences i've had as i've been able to have these conversations with others and share that expertise here all right great and hillary hi everyone i'm hilary mason i'm pretty excited to be joining this discussion today i've been working in machine learning and data science for about 20 years starting first in academia and then transitioning to industry i like to build things starting about 10 years ago i've been primarily in management roles um leading data science teams as a chief scientist as a ceo of my own company as a general manager uh and then i started a new company in the last uh couple months which has been its own interesting experience um but one of the the great joys of uh having been around and being a manager is that i get to help launch some really brilliant careers and uh in this context that means i've been having a lot of really deep uh career conversations with folks across the data spectrum so pretty excited for this conversation today awesome and jacqueline hi uh my name is jacqueline nolas i have a background in mathematics i got undergrad master's in math phd in industrial engineering i've been doing data science for about 15 years now and probably about eight or ten of those as a consultant so working as a data science consultant becoming a director at a data science director at a consulting firm leading a team of data scientists realizing that i didn't like the management tax as much and much more like actually writing my own code and so then working as a freelance consultant for a couple years now but um with the uh virus making consulting independent consulting very difficult i uh last week started a new job at a company called brightloom as a principal data scientist so not only have i uh am i on a panel about advanced degree and the pandemic that i also have um and uh i also uh i recently in last year wrote a book with emily robinson uh build a career in data science all about how to become a data scientist or become a more senior one decide to be a manager um make all those sorts of career decisions awesome great so uh if you're in the audience you can see we've got a bunch of different perspectives here for uh to uh share with us so uh don't hesitate to get your comments in uh via the chat you know i'd like to just start with a a pretty broad question about you know what are each of you seeing and what is kind of different and unique about the environment that folks are facing right now from you know the the particular perspectives that you're coming at things we can start with you carolyn i'm not going to use the word unprecedented but it's tempting in europe as much as i have noticed um there was a decrease a significant one of the job position being published online meaning on linkedin and different job boards so it was a 60 decrease for instance in france um so there are two ways of analyzing it yes there was less than usual but there was still 40 of job offers being out there so i would say that was different for from a candidate perspective is that the the the number of opportunities was significantly less so i could notice that candidates were a little bit more creative in the way they communicated online and the way they gave visibility to their work so for instance i saw people getting back to the recruiters that wrote to them in the past and that they ignored and so it's interesting to see a switch from candidates not answering to recruiters but now they go back um and then i think i would like other panelists to to speak about that because i also got the opportunity to help data scientists looking for a job so i can also share with you the tips i gave to them awesome and how about you what are you seeing i think some of the main challenges that i'm seeing is the fact that a lot of uh growth plans have been put on hold so you know while you know we work with a lot of clients around the world and it's organizations that had significant plans to grow this year and hire additional data scientists to their teams are not necessarily saying this is not going to happen but maybe not right now so so i think that has a significant impact right because it's been reduced as caroline was saying and has been reduced the number of jobs out there have been reduced significantly i do think however that this is a temporary effect so so everything that brought the data science careers to the forefront and made it such a high demand job those things still exist out there i think it's just everyone is going through this adjustment process so uh if anything for you know from the perspective of people out there looking for new jobs it just means that they need to do more of what we're doing and and really focus on doing it better and and more being able to sell themselves better because the opportunities are going to be reduced at least you know for for for a period of time that we don't know how long that's going to be jacqueline you mentioned in our pre-chat that a lot of the process of finding your new position was the same but it was just much harder yeah how much harder elaborate on you know what you experienced so um years ago when i was looking for a job i actually tweeted you know hi i'm a data scientist looking in seattle area and i you know i got a certain number of like views of that tweet i got a certain number of leads and i tried doing that again about a month ago a month and a half ago and i got five times as many views and maybe a fifth the leads so like basically a couple years ago when i did it it's like oh what about this company this gun and then this time i was like have you tried amazon like it was just nothing and i was just astounded because my my resume had saved largely the same if anything it presumably got better um but it was just like it was just total silence um and you know i was able to get a job and i'm very happy with this new job very excited about it but just the number of opportunities is just minuscule compared to what it was a few years ago how about you hillary you're talking to a bunch of folks out there what kind of experiences are they having yeah i mean maybe i'll take the perspective here of people who are hiring um because i've spent a fair bit of time focusing on building teams i have a lot of friends and colleagues who are in that position and i think it speaks to the of course there are fewer positions open but the types of positions that will remain open will not be the ones where it's like let's hire some data people and throw them at the wall and see what they managed to make for us the ones that are open are the people who have a business generally at some scale where there is something to optimize and they know they need a certain set of skills to do that and those folks are gleeful about some of the talent coming on the market um so it's not you know all doom and gloom i'll just say from where i'm sitting i'm seeing from that perspective the landscape of opportunity that exists has changed pretty dramatically in the direction and there have always been a bunch of flavors of data science jobs some very research oriented some very you know you optimize one metric and that is all you do um many in between uh many you know in the the sort of you're the bridge from the the data and tech into a business unit like maybe you're working with financial um reporting um you'll see more more jobs that are a clear need where there's you know revenue and growth already attached and fewer of the highly speculative fewer research type positions right now and that's from the the folks who are hiring and creating those positions um so anna brought up an interesting point that uh i'll segue into one of the questions that we've gotten uh and that is that we're you know immediately prior to the pandemic was a period where you know data scientist was an extreme growth you know area growth opportunity growth career and as a result of that we've had lots of people joining training programs boot camps over the past few years traditional educational institutions have spun up data science programs so we've got not just folks that are you know experienced and and in positions impacted by layoffs but new data scientists and so we've got a question uh from willie costello you know what's your advice for new data scientists that are trying to break into uh the field for the first time anna so i talked to a lot of people in that same situation right they they've done the work of going back to school joining a book camp taking different online courses and really focused on on getting the skills to go into a job of data science my main advice would be do not stay there so one thing that i see that's very common is people that come to me with i've got this certificate or this this course my next question is always going to be how have you used those skills you know have you been able to bring them into your current job have you been able to come up with um topics or interests you and you know what have you done with the skills so my main advice will be as you work on acquiring the actual skills sort of from the learning perspective find ways to put those into practice because i think especially in a in an environment right now where the opportunities are limited we will prioritize people that kind of bring that element of not and it's not just the the sort of official formal real world projects but also the drive they they i did not just you know took take these classes but i just went in and came up with this interesting hypothesis and i built something around it right like all these like cool research innovative data science jobs they're really um they're the ones getting hit the hardest right the jobs that are like the jobs that a lot of them are staying around are the analyst jobs right and i think as a data scientist or if you want to get into a science field you it's often you lost some serious attitude like i'll accept nothing less than the title of data scientist but there are jobs that are not quite title data science but use a lot of the same techniques skills and are great building blocks to help you future your future grow your career and so especially the time like this where things are so volatile um finding a job that like meets the criteria of a lot of the things of data science uses numbers you have to play with num you get to play with data things like that but maybe isn't exactly what you want to do like that is so worthwhile um in a way that wasn't true six months ago on the the topic of building a portfolio and i'm going to come back to your uh your point in a minute jacqueline but on the the returning to the topic of building a portfolio uh myrtle janeau asks what kind of projects would you expect to see or want to see in a portfolio anna one of the things before i talk about the what type of projects i think it's very important to have your your github repository very much up to date so that's one thing that we usually look at and and i know you know when we're working on a resume we're more focused on what we put on the resume but a lot of times if you're not adding your github account this automatic systems can actually even discount your resume altogether before before they even bring you to a recruiter in our case at ibm because we everything that we do is so open source driven so i personally look for projects uh leveraging python because that's where a lot of our clients are working with technology so so i want to see people that have been able to uh really work across the whole you know cycle of building a hypothesis and preparing data and and building some models and isolating your best model using python but i there's something that i always look for that that goes beyond that technical capabilities and it's i like to be able to infer your curiosity levels and that's something that for me is very important because i find that people with very high curiosity levels tend to be really solid data scientists and and really are able to deliver a lot of business value so so it's not so much the you know i use this technique or i use this other technique i i like to see the process i like to see how you thought through uh a question that you had a hypothesis that you had and how you uh brought that to something that would allow someone to bake to make a better decision around that topic i can jump in with one observation as someone who's done quite a lot of hiring requiring people to put things on github is really difficult because certain people are open to harassment if they do that just due to the nature of who they are and so i'd really encourage folks to take a step back and say what do we want to get out of seeing a candidate that has written code in public um and are we going to provide other ways for them to demonstrate that they are capable of writing code uh that they accomplish things that they have their name on things um and so i'd really encourage folks especially people hiring to be very thoughtful about what you require on those applications because you will be excluding a fair number of people from your candidate pool or forcing them to take on some kind of risk that really isn't appropriate or evenly distributed but then i'd also like to reinforce the point about creativity and curiosity and i think that in so many discussions of data science we focus on the technical capabilities like do you know pytorch you know python um but actually what makes a great professional is that curiosity and creativity and we assume you're going to be able to learn what you need to learn as you go and like frankly nothing that i use today existed when i was in school um and so really looking for folks who have the ability to know when they should start to learn something new too awesome carolyn did you have something to add i agree with both anna and hilary um my advice is always to harmonize everything that is going to present your background and your work meaning i don't want to find different information whether it's on linkedin or github or kaggle so first no matter if you're a junior person jumping into data science or senior person try to look for harmony in what you describe about you and i agree with what hillary said because there are biases and we should not expect all candidates to have public repo repositories however i always encourage candidates to try to build the bridges between the expectation of a company you know look at their blog post see what kind of technical issues they face the technologies they use and even if you have a different kind of background try to narrow the the difference and make it easier for a recruiter to point out why you are a relevant candidate it can be curiosity it can be the time you spend on training yourself it can be you know i was teaching kids how to code this stuff for this kind of stuff we all need to value what we do and just to build a little bit further on all these really great points i mean i think it's just like just be passionate about it like one cool blog post or like one i saw like one time i hired a person because she made a really cool power bi dashboard about a theme park and it's just like i love that stuff right like just one small interesting thing that's going to hook me in does so much better than a jupiter notebook with 10 000 likes of python showing that you use tensorflow i guess but i don't know because i can't really follow what's going on in it like just if you are gonna put something out there just make it like make it something someone's you know gonna be a little intrigued by um and like that involves like really just like being curious as was discussed but like yeah just like add some zest like make it about you jacqueline did you find that the portfolio what played as a bigger role in your recent job search as a more senior candidate um you know it was fun in interviews to be like oh actually i wrote a book um that was fun um the thing that so here's the interesting thing i don't think very often that like necessarily they're like oh i looked at your github page and i saw this package and you know i looked at line 27 and i saw you really did something clever there so i'm gonna hire you it wasn't like that but what was happened was like they loosely saw oh you'd have some cursing like the cursory size and knowledge on this like i like for one of my side projects one time i uh made a neural network that i trained to generate offensive license plates um using a historic data set of arizona offensive license plate and it was like i never used a neural network before and this taught me neural nets and then i went on to implement that at like a massive company in like a huge ml pipeline and i only learned it because of that silly side project and so what did the company do with all those offensive licenses that you generated like just this knowledge you know just learning this stuff really helps and them seeing that like oh you clearly like i gave a talk about this and that caused that you know like like just seeing the excitement build really goes a lot farther than showing that you know a particular technique or a particular couple lines of clever code so we've talked uh about things for people new to data science we've talked about folks that have experience carolyn have you run into folks that are career switching so they've got experience in some career but are trying to transition into data science and do you have any unique perspective for them yes indeed and on this topic i have a lot of uh beautiful stories thanks to the umlds chapter in paris because um for instance we chose to run our meetups only in english to be inclusive to foreigners joining paris and we have helped astrophysicists mathematicians wanting to switch from their initial field into data science so i would really encourage people to connect with the support group it can be our ladies we mlds women who code um even the online platforms such as twitter kaggle sometimes stack overflow can be useful on this topic as well but you know to connect to people who might have done the same switch as you to first know that people have done it so you can do it to get also insights of where are these people working so it means that possibly these companies could be hiring you and i really believe in the strength of network offline and online and right now with the pandemic you have to make it as much as much as possible online so these support groups are existing so if you want one example because you know i don't want to just give ideas i want to give working ideas we had a mathematician she was american and she came to paris and you know she didn't know where to send her application so one of our members referred her in her company and at the end of the day she got hired and the thing is she asked for help in order to prepare herself for the phone interview with the recruiter because it was the first time for her that she was going to present her background for a new job that she hasn't done so this is something ask for support and don't try to remain isolated one of the topics that came up in our our pre-chat was the idea that you know there's a difference between a data science job and a good data science job and that uh as anxious as we all are about you know unprecedented times and um you know as important as as jobs are that you know there's uh if you can be a little bit more discriminating in the way you pursue opportunities that will have long-term benefits for you i think it was you hillary who originally brought that point up can you elaborate on it absolutely um one of the things that i find somewhat astounding still about data science is that if you go read data scientist job descriptions at a bunch of different companies they're all different um and you'll realize like once you start looking at these positions that they actually uh work really differently so some are centralized teams sometimes you're the only data scientist on your team sometimes you're in a group that's responsible for operational analytics and making some predictions but you're not doing machine learning sometimes you're in a research team uh you sit in different places in the org you have different types of managers and management experiences um and then you layer on top of that all of the complexity of choosing a company to work at any way so you know is it a supportive culture a competitive culture a political culture um do they really invest in their own people and grow their careers or is it more of a burn them out churn them out sort of place um so there are a lot of things to consider when you think about what's a data science job and what's a good data science job and what's the best data science job for you um and it is always hard to find those good and best jobs under any circumstances much less now and so i'd really encourage folks to be very thoughtful and ask a lot of questions about you know how does this team work together am i going to be the only data scientist on my team and by the way if you're entering your career try to find a place where you can work with a team of people who have been doing it for a while because your first job is going to set you up for your second job and your third job and the more you learn from people with experience the faster you'll be able to grow not as much a concern if you're very senior and you're quite happy being the only data scientist on the team um so it's really they're all of these things to be thoughtful about even beyond what is my title what is my salary you know sort of how do we work together and i think we don't in our community we don't really talk about those things nearly enough can and on that point um it's gonna take you a while before you figure out what are the things you want in the job like i think the first seven years of my career i was switching jobs because like no i don't like this part no i don't like this part no i don't like this part like it really took me a long time before i really got comfortable with like okay here are the things i have to have in a job your things that are nice to have the things i don't care about and like for me now this bar in my career like the thing that i need more than anything is if i'm like hey this is a good idea or hey this is a bad idea i actually have a chance of influencing the like the trajectory and sometimes i got a big company you may not get that so that's something that's really important to me but i know lots of data scientists who like if they just get to be in their corner doing their data science they do not care if they get to influence the like long-term term trajectory of stuff so i think that point's totally right i think it's fine like if you get into a job and it turns out you don't like something about it that is not a failure of you that is a very normal natural part of this process and like it's just a it's a whole growing process and these things change too as you get more senior and your life goes on what your priorities are may change too i also like to think about it in terms of the things you hate less than other people hate them so like maybe you may be at a point in your life where you don't mind working 60 to 80 hours a week you may be at a point where you don't mind travel once we're allowed to do that again but you may be politics doesn't bother you a bit or that you're great at office politics um and so it's really figuring out yeah what flavor of company and culture is gonna be one where you can drive and and i would add to that hillary what flavor of a job in the data science ecosystem because this is this is not you know i think years ago let's say let's use another example you were a database administrator that was your job and it was very clear and i think when it comes to the field of data science machine learning artificial intelligence it's all this ecosystem of people that may have different roles at different stages throughout the build out of the solution so you know you have business analysts and you have developers that are maybe consuming the output of a model and integrated into an application you have people that are managing these models in production and so so i think a lot of one recommendation that i would give to people in the audience is explore all these sort of adjacent roles do do not necessarily go with a job title when you're applying for a job because as hillary said you you put 10 companies and 10 job descriptions for data scientists and i we can assure you that they're all going to be different so so do not go just by by just a job title break that down into what it entails what a job entails same with your skills instead of saying oh i'm a data scientist or i'm this or i'm that really break it down to the skills that you have and and bring those forward in a in a conversation and an interview because i think at the at the end the most important thing is what can you do what are you able to accomplish and and that uh the the name for that umbrella may be different all over the place could i have one quick thing which is to say also i think we also have a notion that you have to progress on a ladder like you have to be a junior data scientist and a data scientist and a senior data scientist and then maybe you face the existential crisis crisis of like management and that philosophical uh change in your role and at least i may be a bad role model but i've been a data scientist a software engineer a ceo a corporate executive the world's shitty assisted men like you can bounce back and forth uh and do this pretty successfully so look at your contribution yeah and i really don't think i think that's true and i also don't think that one is necessarily better than the other i think for some reason there's a stigma that like machine learning engineering is superior to decision science or data science is superior to um analysts and like there's a lot of these con preconceived notions that are not true and also like if you if you dig into them a little bit they're really messed up um so like don't let the idea that like oh i i am i'm not that job is beneath me because it's decision science or i can't do that job because i'm only a decision scientist like don't let that stuff hold you back caroline i've got to imagine that as you work with candidates you are faced with you know folks that aren't sure exactly what they want to do or folks that have lots of different job offers and need to choose between them is there a you know a framework that folks should you know keep in mind as they're trying to uh you know either figure out what they want to do or evaluate different opportunities or point themselves in the right direction career-wise good question first of all i think good candidates are informed and prepared candidates meaning i will second what hillary said earlier before starting to apply to anything have a look at the salary ranges the companies hiring in the field you're targeting i don't know if you have been enjoying psychic learn or numpy and you want to keep working on these libraries look at the kind of applications and companies using those look at the experts in the field and maybe try to connect with them because right now they have much more time to dedicate to peers aspiring data scientists to talk about their job and how you know they built their own careers and usually you know it's a recruiter technique um don't write it to someone saying hey i'm looking for a job you can do that if you are if you're okay but i think it's too aggressive but write to someone hey i am wondering if would you mind sparing me 15 minutes on a zoom or hangout call and try you know to understand what you would be enjoying on a daily basis so this applies to junior people and then if you are experienced the thing is um i'm always thinking about this uh article the engineer manager pendulum from charity myers you know the more you grow in the management ladder the more it's going to be narrow for the to look for your next step so while you know that get ready for describing your technical projects get ready to talk about your aspirations get ready to talk about the technical issues you faced yes i was modeling this algorithm it was kicking ass on the paper and then my my data was noisy so here's the solution i found prepare that as much as possible because no one can do that better than yourself especially if you have a recruiter in between you and the hiring managers and the future team members this might be a little bit less valid if you directly speak with a peer so that being said ask a colleague that you trust to make a fake interview with you to evaluate how much you need to get ready or you can list questions and ask for a member of your family to ask them to you and you hear yourself you correct yourself and you become ready when you know you want to do great as an interview so these are pragmatic pieces of advice i would like to to point because um the crisis stressed something we are going to prioritize the candidates and the hunt is even more difficult than before i want to take a few minutes to jump into i mean i've been following pulling questions in from from the chat throughout but i want to jump into a few of them a lot of the questions are of the um hey do i need an x you know and an x might be a phd or a kaggle rating or um a portfolio we've talked quite a bit about that you know any specific responses to you know phd or kaggle ratings or i'll tackle i'll talk about those so i have a phd phd not good idea don't get it i mean not if you want to go be a data scientist in industry um just because like oh my god it is so many years of your life it is so much lost income and the percentage of stuff you learn that's actually useful to working in industry is very small there is some value you learn how to learn you learn how to kind of like do some portfolio building when making a dissertation but nothing compared to actually the path of going into industry just before a phd which is not to say that no one should ever get a phd or the phd won't ever help you but just if your question is like i really want to go into data science i don't know if i need a phd or not like no no don't beware um and then the kaggle score um this is kind of this is like it's like the same thing with the github account where it's like it comes this like weird metric where like oh you need account you want to apply you don't even have a kaggle score right you don't even have a github and it's like no you don't need these things these things can be helpful right being able to say like look i did this kaggle competition i did quite well i can make a story around it that could be helpful in an interview just like a github portfolio can be helpful in like look here's some things i can show you how you can did it but there are a thousand ways you can show a person that you are capable of something and the idea that you have to use a preset thing right like you could right if you go to a conference you give a talk you can make a poster you can make a cool website just for you there's like a thousand things you can do that are show and interview you have more capability than just what shows on a resume and none of them have to fall into github kaggle doctorate um and don't don't can can i add one thing to that there's one thing that i actually find very useful and it's for someone to have their elevator pitch so if you only have one minute with me if you come up you know and send me an email on linkedin for example tell me why i you know i should be interested in talking to you tell me what you think is the value that you bring to the table and and that can be done in you know literally a few sentences and it's much harder to do than it sounds it's actually much harder to do than if i tell you talk to me for an hour about your life story because you really need to bring it down to the most important things but i but i do find that that's very valuable as you know of course now we're all like locked in our houses but if you're in a up and you can talk to someone you can tell them hey this is what i do this is the value that i bring to the table i find that that every person should have their elevator pitch ready uh when they have that opportunity via digital via a face-to-face meeting you know whatever that is and i know we sometimes don't spend enough time thinking about that because we spend more time thinking about the project and the technical skills and all that and it's like summarize to me your value in a few sentences and i can guarantee that that will get you you know doors open uh because it'll be easy to understand that's a nice way to just sort of put a framework on this question which is do i need an x is usually about opening the door there's how you get through the door which is how you get through an interview process and then there's how you actually are amazing at the job these are three completely different things correct um and any of these x things can help you open a door but there are a lot of ways to open a door that don't involve having a kaggle score um don't involve necessarily having a certain degree of any kind um in fact with kaggle stuff i personally um and way more interested in the questions someone is asking themselves so it's cool if you've done a bunch of kaggle competitions but they give you the question i want to see you know maybe uh you know i did this kaggle competition and then i got really curious about this question and then i found this other data set and i took it off in this direction um so it is uh you know whatever you find to open the door um and that could be a good elevator pitch it could be a twitter thread it could be a visualization you make um there are lots of ways to open doors and there's no one right way and there are certainly some that cost a lot more in terms of time and effort than others and i think when it comes to making an elevator pitch like i remember when i was first finishing like first going to get my first job thinking like like i don't know how to make an elevator pitch like there's nothing special about me i got a degree and i want a job doing this stuff like what is there like how do i say and i think this is where that portfolio goes a long way because by like reading doing side projects reading they're checking twitter like there's like you start to kind of understand like what's out there what is more interesting in you and just being like look i'm a data scientist i just finished this bootcamp i'm really you know i'm fascinated by time series analysis um and i'm looking for a job like just having anything like like one like like like kernel of something interesting about you is just a huge difference yes yeah i'll add to the one of the ideas that i found really helpful uh in thinking about careers is something that i picked up from scott adams the creator of the dilbert cartoon he talks about this idea of talent stacking which is essentially you know if you want to be successful you can take the route of hey i'm going to be in the top one percent of whatever field you're going after or you can be pretty good at a couple of things and so his example is you know he's not a great cartoonist he's not a great writer but he's pretty good at both and he mixed them together and he created this amazing thing uh dilbert and there are tons of you know skills and talents out there that you can remix to create a unique um you know unique positioning for yourself caroline you had something to add yeah i am a massive twitter lover and you know sometimes even if you don't know how to talk about yourself you can highlight other people and get visibility about your work and i completely agree with jacqueline like there's plenty of ways of being creative and talking about your work when people write blog posts or now you know the younger data scientist they do tutorials on youtube i think it's great because you know it's not just you show off about yourself you you share knowledge with other and this is how um i think you can get attention so please be vocal on twitter on discord on the different slack accounts you can join and don't forget to give visibility to your work and yourself and if you don't want to promote your own work make a network of people who all can promote each other's work because there is nothing more fun than promoting your friend's work any of you have recommendations for folks that want to take this opportunity to kind of skill up and you know better position themselves for their next thing is there a genetic why could i do could i do just the opposite right so like i was a consultant i was working crazy hours uh the consulting contracts went down and i was waiting you know to my next job that absolutely could have been a time where i learned all these new skills like i'm even better but i could tell that in fact i've been on the verge of burnout for years and so i didn't do that and i used this trying time to make sure that i just stayed mentally and emotionally healthy and really trying to handle that rather than trying to like see if there's just one more skill i could get and i think that ended up being really helpful for me in a way like taking hour long walks was probably more helpful than an hour of learning new tensorflow stuff um which is not to say that leveling up your skills and like that stuff's not great like definitely if you want to do that that's good but also the idea that you have to use the quarantine to do this um i think can be really difficult um and maybe don't do that yeah i i agree with um jack let's not all think that we should be uh productive but my mom used to say um try to transform the negative into something positive so personally you know i was a bit depressed of not going to the gym not being able to go out and so i was like what are the things that i usually push back because i have something more important to do and i did these things and when you are a data scientist it can be updating your different platform it can be reaching out to people it can be joining a platform because you thought you were not good it can be learning how to put code on github because i know that women for instance are underrepresented on github for the reason we mentioned earlier so i like to try to transform something bad into something that you can be proud about later and it can be a work outside well and i think to that point like a lot of things are not data science skills but still useful like i updated the html on my website during the quarantine and like that's like learning more html isn't like data science skills but like that actually helps me when i eventually make a dashboard things like that so there's a broad set of things that could be useful anna how about you you're a lot of your role focuses on skills development any takes on this one yeah so so for me the uh so i get approached a lot by people both internally at ibm and externally that tell me would you spend some time with me i want to become a data scientist can you give me some guidance and and when i do that i kind of flip the conversations to not just telling them this is how you become a data scientist but tell me about you tell me why you're interested in becoming a data scientist so this is an advice that i would give to people in the audience too which is you know break it down into the things that you love and a lot of times i find you know i would say more often than not as we start kind of breaking apart kind of the skills that you need to become a data scientist do you know machine learning i find that a lot of them are really not into that they just felt attracted to a piece of it so i'll give you an example some people tell me oof i could spend my day cleaning data i love to clean data and i love to you know just bring data sets together and so that's more the skill of let's say a data engineer persona and i won't call it a title i'll call it like a persona so if that's your passion you don't need to go and learn all day you know go back and refresh all your math and and really learn machine learning because that's not what you're interested in in other cases i find that people are just interested in developing really cool applications so they're a developer now and they feel like maybe i need to become a data scientist in order to build these really cool applications well maybe not you know maybe if you're not building a machine learning model from scratch if you're really your end goal is to build some cool applications using maybe some pre-built models maybe some pre-built technology focus on that so so i would my advice to people in the audience would be try to break things apart to step away from the data science as a title and kind of bring it down to what are the elements of data science that i'm interested in and maybe they're shortcuts to it to where you really want to be that that are not necessarily uh becoming a data scientist joseph asks um he's run into widespread bias towards older applicants but certainly there are other groups that are biased against in the the job process you know any comment or advice around overcoming those kinds of biases that's a really hard hard drive someone that's a female hispanic and on the older side like i have three strikes right so it is it is challenging and i i personally don't think there's like a magic wand to get through it um i personally focus on my skills the things that i can do the things that i'm really good at uh but that's easier saying that said than done unfortunately if you encounter a situation where you feel like there is biased against you for whatever reason don't waste time on it and most importantly don't let it don't take it personally so just you know there are many jobs out there you only need one of them just keep going and also share information about that employer with other people in your network so they don't waste their time um but you can't fix someone else's bias certainly not as a candidate how about you caroline yes um hillary for the win i want to say um i would add that if you are coming with an internet intersectional background meaning you face different discrimination because you're a black woman transgender um and you face this issue i agree with them i really don't go however there are like groups existing that can help you again to select the right places for instance there is queer and ai blackened ai latin in ai lgbt in tech for instance so i strongly encourage people to go and look for support um because you know um i am a lesbian myself and sometimes you're like no i'm totally imagining something and then you discuss with someone facing similar issues and you realize how much it's not just an emotion it's a fact and um if you don't want to work with this kind of people so other people made a work prior to you and go and see who supports the which companies support these communities and who are the people in these communities that can refer you or talk with you to reassure you refer you for the the people who are more privileged listening to this right now um so i started on the my career path as a white man and now i meet trans women and so the be like one of the biggest things now because this is the first time i've been on the job market as a trans woman and the thing that's so different is now when i'm at a job interview i have to spend so much effort being like are these people gonna be good to me when i'm like when i'm here like how many women are there like i'm doing all this mental math that i just did not have to do as a guy i just like was like oh cool this job seems nice i like the tech i'm here and that burden is real it's on like minorities all the time and it is the worst um so it's just like this is like i totally the person asked the question what you're dealing with is totally valid everyone who's in these situations is totally valid if you're like oh that's not real um because i haven't had to deal with it that probably means you have not been on the receiving end of it but it is definitely valid and real yeah i was just going to say that in this field diversity is so important so so one of the things that i that i would encourage you as a job seeker is to research the companies that you want to work for because there's definitely a difference i mean some companies are really more focused and more intentional about building diversity in general but especially in data science teams because we don't want a team of the same people you know uh kind of driving decision making in the organization or influence influencing significantly decision-making in the organization so very important to research the companies because you will find the difference in in terms of their diversity and inclusivity efforts i want to make sure we get to a question uh by juicer having done a bachelor in statistics data science seems best for him but a lot of universities are adding a master's in cs prerequisite uh so what's actually the most important aspect of positioning yourself we've kind of come at this from different directions is there an ideal set of skills or benefits so i will say this um i started my career long before data science even existed as a term i think most of us here are judging by everyone's years on the field and so for companies for years had to grapple with the fact that there's no set list requirement for what is a data scientist and that companies we're just starting to get to the point where you can get a degree in data science but it's nowhere near normalized like that is not standard at all and like some company you know some schools yeah maybe they require cs degree undergrad maybe they don't but if they are putting these strict requirements in there they're making big assumptions that are probably not true like the fact that you need a cs undergrad to do masters then data science is like requests um and so and if similarly if a company is like oh we can't hire you you only have a stats degree to do data science that's mind-boggling so just you know just like if the companies are doing this that is more of a sign that they do not know what they actually want on a team and they're probably not a team you'd work want to work for then it is a sign of like oh you really did not take the right four year path or the 15 years of tensorflow experience yeah come on what about the the process during kind of shelter in place we've talked a little bit about how networking is harder interviews are different now any tips for managing that caroline what are you seeing in terms of the way the process has shifted uh today yeah so massively you know companies that were usually reluctant to have skype interviews or video oriented interviews are now completely open surprisingly and this is also adding another layer of difficulty for candidates because you know when you interact with someone the body language counts a lot of the percentage of how convincing you're going to look and sound so it might sound super pragmatic but while going through offline interviews make sure to look at the camera make sure to have a working microphone make sure to have something to take notes in order to keep up with the flow of the conversation these might sound silly but you know i've had like people you know like that and then you have the gardener coming with the in the back in the background this is not possible this is not professional if it's a kid it's okay but anyway this is different and i wanted also to suggest something that we mentioned with meetups luckily meetup groups are going also online so if you want i would invite people to reach out to meetups organizer to try possibly to introduce their work because it might be a little bit less impressive to present it while you're at home comfortably seated instead of having like uh hundreds of faces in front of you so this is another negative thing that you can switch into an opportunity and you thoughts on taking the process virtual i would just say to that point um the lack of conferences is really hard but like twitter data science twitter is so good like i really get a lot of value meeting people replying to them and it's just just twitter that's the social network for me um so i'd highly recommend if you're a data scientist you want a network you can't go to conferences and you're not on twitter then like definitely get on twitter awesome awesome so we're uh winding now now as the last few questions come in um maybe i'll have each of you take a moment to kind of your top three words of advice or top couple of you know takeaways the number one thing you would do whichever way you'd like to go with it uh hillary great um i think my top takeaway or a bit of advice is that um at least a lot of the folks i've been talking to now and my friends who are job hunting are doing so in an environment that has a ton of anxiety uh in it and of course making major life decisions while you're suffering through anxiety and uh you know stuck at home dealing with responsibilities there is always a terrible idea so the one bit of advice i'd say is to try to keep things as systematic and objective as you can to remember that you are awesome there are a lot of people out there who will help celebrate you sam being the one who kicked all of this off by retweeting people's accomplishments i'm happy to do the same um and uh you need to set down some criteria and rules for yourself about you know what you need to make on top of that what are the things that you prioritize in a position that'll make it one where you can actually really succeed over time um and then systematically approach finding it and trying to do so in a way that your own anxiety doesn't drive you to weaker choices than you might have otherwise made and i know that's very hard to do so rely on your friends to help you there too carolyn yeah on my side i would say book discussions with peers read watch tutorials make something that is going to inspire you and i'm going to quote jacqueline walk outside if it's your thing and i wanted also to stress successful stories um i have helped a student an indian student who got you know uh abandoned he was supposed to do an internship in amsterdam and he reached out to me and i just connected him with my indian network and he just uh found um an internship another female member of umlds reached out because she was looking for a phd she did a linkedin post a twitter post and she got hundreds of answers so don't give up just try stuff and if it doesn't work it's okay it's like a massive a b test that we are all going through and we are going to learn the way echelon yeah so i think what i'm going to say is like kind of a rewording what hillary said like i really believe it um you know so i just got off the job market because i was looking for a new job and i really like i had to make a lot of like like letting go ego things so i'm very happy with the job i got principal data scientist like small company like really was looking for but like i had to like really like i was checking linkedin and i'd like say like i guess i'm ready to take a job at like a lower title that i'm used to and like i have to give up like that's a lot of career progression i was giving up when i was making those decisions and it's just terrible and there's not like a cool thing i could say on a panel that makes that easy and it's not like there's anything about me that made it happen it was the environment it was the virus but that's the reality of the situation and i think the more you can dissociate that from your own well like sense of worth like the more i can be like i'm still going to be good to see a scientist even i'm still going to love it even if i'm a senior data scientist instead of a principle things like that the easier it is but that's not something that's easy to do and it takes a lot of practice and time and like you shouldn't beat yourself up over it if you are having these sorts of struggles anna um what i would say is that this is a multi-disciplinary field so as you are out there let's say you're transitioning from one career to a data science career or just embarking on it make sure that you really sell all the other expertise that you have as it relates to data design so i'll just use an example you worked in marketing for 10 years and you've been upskilling yourself and you you know you came up with a couple of projects in your company but you've only been doing data science for a year um i would say your value is it's exponentially higher when you bring together all the marketing expertise and your data science knowledge so a lot of times in our efforts to switch careers we want to oversell the new thing and say i've been doing all this and kind of ignored the rest i would say in this field at least from my perspective because it's such a multidisciplinary field because having different perspectives different domain expertise is so critical make sure that you are not underselling your whole package of skills one meta question we've gotten we've we've had a bunch of questions come in i tried to identify the ones that have the most broad impact but a lot of people have very specific questions i'm in situation x i'm thinking about why what do i do any thoughts on how uh folks can get those kinds of questions answered sure i'll take a stab google it follow people who are good on twitter do some research and if it turns out that no one has a consistent answer there's not a quick answer there's probably no answer right like the question i was just saying should i drop down to a senior data scientist because of the virus like how will that impact my community no one's gonna know that yeah like i don't know that like are you gonna know that sam like no one's gonna know that and i think the idea that each one of these questions needs an answer before you could proceed on with your life is like that like let go of the idea that you have to have a definitive answer do some googling follow experts um read blog posts but don't like don't try and like optimize every little moment of your life yeah i think that echoes a point that hillary made in our uh our pre-chat um you know this is a time of tremendous uncertainty and we don't you know we don't know what's coming we've not experienced this before uh and so to a large degree kind of pursuing a career moves in this time is going to require being at least uh being comfortable with that certainty uncertainty at least to uh some degree oh i also should say oh and buy my book that'll get you the answers that's the one way that's it awesome well uh i'd like to thank all of you for uh participating in this panel what a wonderful discussion and thanks everyone for contributing your questions in in the chat once again thanks to ibm for sponsoring this discussion we will have the video well the video will be up on youtube as soon as we're done at the same url so be sure to share it with your friends and you know feel free to continue the conversation in the comments thanks so much for participating everyone thank you all right everyone that's our show for today for more information on today's show visit twimlayi.com shows as always thanks so much for listening and catch you next time wow you\n"