Joseph Powers @ Intuit & Jen Wang | Data Science Hangout

Well, hey, everybody. Welcome to the Data Science Hangout. If this is your first time joining us today, hello. It's so nice to meet you. I'm Rachel. I lead Customer Marketing at Posit. If this is your first Data Science Hangout, say hi in the chat, too, because we all want to welcome you in here, especially anybody joining for the first time. This is our open space to chat about data science leadership, questions you're facing, and getting to hear about what's going on in the world of data across all different industries. We're here every Thursday at the same time, same place. So if you are watching on YouTube later sometime in the future, and you want to join us live, you can use the link below to add it to your calendar.

At the Hangouts, we're all dedicated to making this a welcoming environment for everybody. Love to hear from everyone, no matter your years of experience, your titles, your languages you work in, or industry. And it's totally okay if you just want to listen in here. Maybe this is your lunch break, or you're out for a walk or something. But if you want to jump in and add some context to something or ask questions, you can raise your hand on Zoom, and I'll keep an eye out. You could put questions in the Zoom chat. And feel free to just put a little star next to it if it's something you want me to read out loud instead. And then lastly, we also have a Slido link where you can ask questions anonymously, which I realize I forgot to create. So I will create that in one of the first sessions, and then we'll share it in the chat here.

But with that, I am so excited to introduce my two co-hosts for today, Joseph Powers, Principal Data Scientist at Intuit, and Jen Wang, formerly VP of Product and Growth at ThredUP. And I thought today was a really fun example of how community and data friends happen. I met Joe at the Posit conference recently, and we had a bit of an impromptu pizza party with Javier and Sep from the Hangouts here too. And then Joe also introduced me to Jen here as well. So I'd love to have you both introduce yourself and share a little bit about your background. Also, something you like to do outside of work too.

Jen, you should lead off.

Oh, no. Okay, um, so I'm Jen Wang. I, yeah, was at ThredUP, which is an online marketplace for secondhand clothing for five years. And before that, had done a PhD in kind of behavioral science, looking at, you know, environmental decision making. Joe and I met during grad school actually taking a bunch of classes, but one of them being really the data science class that probably turned us both into data scientists. And so I'm really excited to be here and have spent a lot of time thinking about career and all these things is something that Joe and I have talked about over the last many, many years. And then outside of work, I mean, I love to do a lot of things, but most recently I've been cooking a lot, a lot. So I'll just mention that.

Yeah, I had several careers previous to this one. I was initially a furniture maker for seven years, and I started teaching woodworking and became interested in like questions around like students' motivation and how people like perception of their environment can like improve their, their learning outcomes. And so that led to, yeah, getting a PhD in educational psychology, and I had to learn programming and statistics along the way. And these data science jobs emerged in Silicon Valley around the same time. So that was, that brought me here. Outside of work, I still enjoy furniture making. I'm just grateful not to have to rely on it for my primary income anymore. So yeah.

Transitioning from academia to industry

But at the Hangouts, we do often talk a bit about hiring, but it's more often from the employer side, but thinking more from the candidate side. I know you both have some experience transitioning from academia to industry, and would love to just learn a little bit from you about maybe some tips you might have for us or people who are in a similar position.

Joe, I'll let you go first.

Okay. Yeah, I'm selecting on my own path. And then I also still speak with a lot of people regularly who are transitioning. And I'll try to offer the advice that I wish I had, you know, six years ago, which is that like, if you're coming out of an academic background, it's easy to feel like you're not qualified or you're unprepared for tech careers in industry. And that's really, really far from the truth. You're actually, you have an incredibly unique and valued skill set. The challenge is learning how to talk about it in ways that a non-academic is going to recognize. And I hope that's something you get out of today. And so, and this doesn't just apply to academics. You know, if you had a career previously in like nonprofit work or government work, and you know, you're thinking about transitioning to tech, I think a lot of this is going to apply for you too.

And so, what are the big things? Like, I think it would be easy to miss your soft skills. And I don't use that soft term dismissively at all. They are the most valuable skills. And these are things like the ability to take a vaguely scoped problem and turn it into an actionable line of research. If you finish the PhD, you know how to do that. Other factors like your ability to translate your expertise to people who are intelligent, but don't share your particular expertise. That is a huge skill set. You know, in order for you to have influence within your organization, you're going to need to translate your expertise to leaders and other stakeholders who, you know, don't have your particular background. And you've got a lot of practice in public presentation and translation during your academic experience that most people never get.

I was never formally trained in public presentation until I got into my PhD program. And then, just out of like self-preservation, you know, your lab members to protect the, you know, reputation and integrity of the lab, they're going to give you a lot of guided feedback on how to present well. So, it's easy to get hung up on like, oh, I don't know how to do this particular modeling technique in Python. And it's like, those are the hard skills in the sense of like soft skills, but you can learn those pretty quick. You're bringing in a lot of soft skills that are very, very rare. They're so rare, they're not going to be in the job posting a lot of times. You're going to need to proactively plug them throughout your resume, throughout the interview process. And I promise you, you will see the hiring managers perk up when they start hearing about these really rare abilities that you've had the last five, six years to develop.

You're going to need to proactively plug them throughout your resume, throughout the interview process. And I promise you, you will see the hiring managers perk up when they start hearing about these really rare abilities that you've had the last five, six years to develop.

Yeah. And I guess to add on to that, I think there are two things that kind of come to mind for me here is that one is that expertise is truly relative, right? Anybody who works in industry knows this. You easily are the expert on XYZ thing in your organization because as long as you provide, as long as you know more about that than anybody else in your organization, you are the resident expert. And I think that when I was transitioning, that was one thing that I worried about is, oh, I don't know enough. But I was just talking to somebody yesterday and I was telling her like, you already know way more about this industry than anybody I've ever hired. And 10 out of 10 times, I would rather take somebody who knows a little bit less about the industry, but has a proven track record of being able to learn very quickly and can pick it up.

So I think that's something, there's not a very standard, standard path that like, if you just do these things, things will be guaranteed. And that can make it very tricky. It also means that, you know, kind of what Joe was saying, your technical skills are not enough. You have to have this combination of technical skills and the ability to sort of navigate organizations and talk to people and build relationships, because I think it's the combination of those two things that makes people, that actually helps them.

And so, what are the big things? Like, I think it would be easy to miss your soft skills. And I don't use that soft term dismissively at all. They are the most valuable skills. And these are things like the ability to take a vaguely scoped problem and turn it into an actionable line of research. If you finish the PhD, you know how to do that.

And so, you know, if you're, if you're seeking promotion within your company, like my advice would be like, start that as an explicit conversation, you know, with your manager about like, what are the skills that you know, you need to see demonstrated? And so, you know, if you're, if you're seeking promotion within your company, like my advice would be like, start that as an explicit conversation, you know, with your manager about like, what are the skills that you know, you need to see demonstrated?

And I want to jump on top of that one because really it's really good about talking to people. LinkedIn is an incredible tool for making the transition. And maybe I don't know a lot of people in industry yet. I didn't know. I didn't either at one point. And actually, you'll find that cold calling on LinkedIn can be very, very effective. I typically look for people who had like a PhD in their name or who I could see had some academic background or some background similar to mine, and then just reach out and say like, look, I'm trying to make the transition. Could you spare 15 minutes to talk to me? These can be very vulnerable conversations. You're not trying to impress them right now. You're trying, like Jen was saying, to get like a lay of the land and to scope like how your skills might fit in.

And overwhelmingly, those people actually turn into your backers. Like they'll submit your resume internally, really just after an hour meeting. Like they'll see that you're like a thoughtful, resourceful person. And you're a safe bet. But to Jen's point, like you learn the lay of the land so that you're better prepared to talk about your skills and what you have to offer. And also to know for where you'd fit in.

My experience has been that like the job levels within data science are very tied to how much direction versus self-direction you're going to engage in. So when you come in as like a base level data scientist or analyst, your work is very highly scoped for you. It tends to be like execute this solution to oversimplify it. You move up a level and now you're being asked, you know, find a solution to this problem that I'm giving you. You move up another level to staff and now you're being asked choose which problems are worth solving. And so like as you move up your scope and the amount of autonomy and responsibility you have for directing your work, it keeps increasing. But so yeah, I would say like, you know, your experiences, yeah, as you get started, your work tends to be more scoped. But then as you demonstrate that you can choose good solutions and choose problems worth solving, you tend to be given more and more autonomy. And then, you know, the responsibility and promotions go with that.

Yeah, I think the other thing that kind of your question, by the way, I'm loving this chat. There's so much like good sharing from everybody. But one of the other things that I was thinking about, too, is that essentially, right, with any job and any organization, there are tradeoffs. And so I think one of the big tradeoffs with academia, in theory, at least, is that with academia, you're supposed to get like full intellectual freedom, although hard to get until you get tenure. But, you know, in theory, you get full academic freedom and you can kind of pursue and set a lot of your own schedule and your work. And the reason I think you can do that is that your lab or where you work kind of, they act autonomously, right? They're in their own line of research, etc. When you go into a company, especially the bigger the company gets, there are just more interdependencies, right? You're trying to work on things that affect and interact with more people. But the tradeoff for that is that oftentimes the impact and the reach is also very different, right?

Two verticals in data science

I have generally found there are two verticals. There's like a machine learning AI vertical, that tends to be about like automating decision making. And if you're automating small decisions, like what should I put in my cart next? What should I watch next? And then there's an analytics vertical. And that tends to be very thoughtfully constructing big decisions, like which market should we pursue next? You know, which product feature should we launch next? And those are very different. And, you know, you may find that you're more qualified for one or the other. If you have a ton of machine learning and predictive analytics experience, like maybe going down that ML path is going to be really promising for you. If you're really into like theory building and like understanding customer behavior, that analytics vertical can be a fantastic fit.

Because they're not going to call it theory building, but like, what are you doing? Like you're developing customer narratives, you're trying to understand customer behavior. And then you're trying to change that behavior in, you know, hopefully positive ways. And that is very close to the theory building that you did as an academic, they're just, again, translating, they're not going to call it theory in industry, they're going to call it like narrative and context. And so I saw a great comment about can I count my years of PhD and postdoc experience as industry experience? Absolutely. If you can convince the hiring manager that those years were relevant experience, then they count. And in my mind, they count. And you should be very confident in claiming that they count.

Career advice and job searching

Sure. So actually, mine is Joe. Joe and I just got this like years ago. So when I was looking for my first job, I think, Joe, like maybe you and Bella had this heuristic or something, which is like in any job, you can really only optimize for two things, or at least that's like a way of thinking about it. Like you can only optimize for two things and everything else that you get is like bonus or is great that it comes with a package. But once you try to optimize for more than two things, the equation gets like very complicated and the set of jobs that you can take might become too narrow. So an example is if you care about geography, pay, mission, culture, growth, you can list all these values that you have and that you want out of your next job. It probably is like doable to get something where you're mission aligned and get like a salary or comp package that you're happy with. It's probably possible to get a geography and a comp package that you want. It's much harder to try to get all three in that first job or in the current job that you want.

And so I really love this piece of advice because there are a couple underlying principles here. One is that it means that you need to get really clear on what you want, right? And I think that some people, and I've fallen into this myself too, it's like people approach you with a job, you get excited about it, but you actually didn't start from a place where you were really grounded in your own values and what's really important to you. So I think that's a great starting point is really thinking about like what do you want and what's most important to you in this next step. And then the second piece that I really like about this piece of advice is that, and there are many other frameworks, I've heard people be like list your top three priorities and three extra criteria, et cetera, et cetera. But the other thing I like about this is that it really puts me at least in a mindset where careers are not made by one job, right? So just because in that one job you're optimizing maybe say for comp and geography doesn't mean that your entire career will not be focused on culture, mission, and things like that. And I think thinking of your career and your jobs as steps along the way that's building like a larger career portfolio has been really useful for me.

Just because in that one job you're optimizing maybe say for comp and geography doesn't mean that your entire career will not be focused on culture, mission, and things like that. And I think thinking of your career and your jobs as steps along the way that's building like a larger career portfolio has been really useful for me.

I think a really good piece of advice is it's easy to think you need to build a lot of new skills in order to make a transition. And I don't think that's true. I think you should focus on polishing your existing skills rather than building new ones. Once you're in the job, then you can build new ones. But just you have a lot more skills than you realize. And if you just polish those and then like refine them, you know, if you're doing multiple regression in your work, great. But I want you to know multiple regression like all the way down to the foundations before, you know, you don't need to start learning like every machine learning technique in the book. You know, you can get a first job just on multiple regression, but know it really well.

And just tied to that, though, is that doesn't mean that the job you're qualified for now is going to sound exactly like your PhD dissertation. I see a lot of people like overly narrowly define themselves because that's what academia promotes you to do is you're like you get very narrowly defined in your identity and then you look for a job that sounds like that. And I just encourage you to be open minded. You know, think more about like, you know, if you if you have a background in like, you know, science or like diversity and inclusion, you know, yeah, there are job titles that are called like diversity and inclusion officer, and that's open to you. Or, and you could be a product manager and just bring that amazingly unique perspective to the role and be an amazing product manager or a people manager who builds a team where people can thrive. It doesn't need to be your formal job title that maps like exactly to your, you know, your PhD dissertation. You know, you can just bring that perspective to a traditional role and excel at it.

Networking and getting past the resume screen

Applying through the online portal, I never do it. And for the simple reason that even with the great credential, I would routinely get desk rejected. It requires networking, like cold calls on LinkedIn, finding people who are second and third degree contacts and asking for introductions. Now, I will admit that for many people, even for me, that is painful and unwanted activity. But you can flip that around and think of it as your advantage that most people aren't willing to engage in that kind of knocking on doors, essentially, cold calling to network their way in. Now, once you get that referral, now you have a much higher likelihood of actually getting the initial part of the interview started.

But that kind of networking doesn't end at the interview process. That's a big part of the job, too. I mean, once you're internal, so you know, congratulations, you've now been hired and you're doing work. Well, you're still going to need to figure out how to network within your company in order for your projects to get greenlit, for your projects to go into production. That is a skill unto itself, too. And so if I see candidates who are doing that networking in order to get that internal referral, that's a good signal to me that they're going to be willing to do the kind of personal networking when they're inside the company as well. Submitting to the portal and hoping for the best is just such low likelihood. And once you're inside, finishing your project and pushing to GitHub and hoping for the best, that's not going to be enough. You're going to need to figure out whose buy-in do you need, get their feedback in advance. It's like they're completely parallel processes, getting into the company and then performing once you're inside the company.

Academia vs. industry standards

I think that actually points out a good transition point of like, what are the standards you're working to? If you are doing basic research and you're trying to build knowledge, then having this exclusive body of rigorous findings is the goal. That's not the goal at most businesses. Again, if you're doing medical research, then that's different. You actually, there is more of a scientific goal in there. But oftentimes, you think of businesses or often it's more of an investment strategy. And so now it's not about building an exclusive body of knowledge, it's about really optimal stopping. How long should we explore for so that we can exploit that knowledge for maximum gain? And that's a different standard. It's not a lower standard, it's a different goal. And then that goal brings with it a different standard. So that's a change of mindset that you should be prepared to make. It's not going to serve you well to bring over academic standards to an environment where those won't serve you. So yeah, most business settings are going to be more about optimization than they are about building scientific knowledge, although there will be some overlap.

Yeah, and Eugene, your question kind of reminds me of a conversation I had very early on when I first got into industry, which was that, you know, in academia, what we care about is, if you get a result, like, let's say you run an experiment, and, you know, your null result is like nothing, or you're, you know, the kind of the worst case is basically nothing changes, right? Like, that's like, in academia, that's like the worst case. Now you don't have anything that you can report on. But if you get an unexpected result, either way, it's more positive than you thought, or it's more negative than you thought. Now you have an interesting result. And you can talk about it and try to figure out causality and all these things. In business, kind of to Joe's point, if academia is about building knowledge, in business, I say a lot of the times, it's just about reducing risk. So a lot of the time, the standard of decision making is we don't want, actually, a good result would be like that nothing changes, or it's positive. And the bad result is if it makes the outcomes that we care about worse, right? And so that's, that, to me, is such a fundamental change in how you look at it. Because if you're trying to be like, oh, should we roll out this change in our business? Really, what a lot of leaders care about is they just want to make sure that it doesn't make things worse, right? And so either if it's the same or better, then that's great. But that same result in academia, if it's the same, is seen as like a not very valuable result. And I think that that shift is quite fundamental.

In business, kind of to Joe's point, if academia is about building knowledge, in business, I say a lot of the times, it's just about reducing risk.

Seeking promotion

And I think that, you know, promotion is often has a has a social component. Not to say it's popularity contest, it is say like, you need to be well aligned with, you know, your lead your managers and leaders about like, what skills, what demonstrations are going to warrant that promotion. And so, you know, if you're seeking promotion within your company, like my advice would be like, start that as an explicit conversation, you know, with your manager about like, what are the skills that you know, you need to see demonstrated? What are the, you know, accomplishments that, you know, you need to see so that it's not like a mystery that arrives at the end of the year, but it's like an explicit conversation. And then, you know, you don't have to, you don't have to stop there, like, you know, look around at who else is getting promoted, you know, within your org and like, reading between the lines of like, well, what have they accomplished that distinguish them from, you know, the rest of their, their peers.

I think for promotion, one story kind of comes to mind of kind of earlier when I first jumped to industry, which is I was a very process oriented person. You know, I really believe that if you set up good processes, you basically bias all of your outcomes towards everything is kind of I looked at things like very probabilistically. So it's like, I'll set up a good process. That means that, you know, the probability that I get a better outcome is sort of determined in terms of how you make those decisions. And so I used to, I spent my first six months on the job talking all about these new processes I had implemented and how great they were. And I have, I got feedback that was kind of like, we're an outcomes driven organization, Jen, like all this talk of process, like it's not doing it for us. And I remember my first reaction was just sort of like frustration. I was like, come on, the process is so important. But I think it kind of goes to this advice, which is that different organizations will value different things. And every organization is its own little fiefdom with its own set of values. So you want to spend time getting that feedback, figuring out what actually matters and how it's talked about and how it's framed.

I ended up taking a lot of the same things that I used to talk about from a process perspective, and I just kind of framed it and was able to show how it led to outcome. And I would really focus on talking about the outcomes. And it was a total turnaround in terms of how people perceived the contributions that I made, even though it was the same kind of work, right? So I think that's one thing is like, figure out what matters to the org.

So at Intuit, they have this principles community. So they get all of the principle level, like engineers and data scientists together, and we talk. And one thing that came out of that meeting recently that was really interesting was just this theme of willingness to take informed risk. And I think this is actually, in hindsight, a really important theme of, like, promotion is not assured. You're probably going to have to take some informed risks. But I think that that willingness to take a bet on yourself and pursue something that not everyone else is doing can be really important if you're seeking promotion.

Yeah, that's a great example. That wasn't something that was defined for them. Like, no one was saying, oh, this is how you should do your job. Like, they took a totally new approach to how to accomplish their job and saved an enormous amount of time. And I think those kind of examples are, like, great ways to demonstrate that you have the, like, scoping and autonomy abilities that warrant the next level.

Well, I feel like we could all keep talking about this stuff forever. And this hour went by really, really quickly. But thank you so much, Jen and Joe, for joining and sharing your experience with us all. This has been great. Yeah, thank you. Well, the questions have been so thoughtful. I've just been, I've been learning a lot from the questions. I'm like, oh, great questions. It is really fun to read through the chat. So if anybody wants to save the chat, there's those little three dots on the bottom there. And you can just click save chat too. But Jen and Joe, if people want to get connected with you, is LinkedIn the best place? Yeah. Okay. Awesome. Well, thank you all so much. Have a great rest of the day. Thank you, everybody. Bye. Thank you.