How We Hatched: Jennifer Sample, Chief Technology Officer at Empower AI

Speaker A: Welcome to the PEAR program from Hatchpad, the podcast that gives you a front row seat to candid conversations with tech leaders from the startup world. I'm your host, Tim Winkler, the creator of Hatchpad. And I'm your other host, Mike Gruen. Join us each episode as we bring together two guests to dissect topics at the intersection of technology, startups and career growth. Jennifer, welcome to the PEAR program. This is, uh, another bonus episode in a miniseries that we call How We Hatched. Uh, and so today we are thrilled to be joined by Jennifer Sample, the chief technology officer at Empower AI, uh, where she leads the company's technology vision and execution to support federal government missions. Uh, Jennifer, your career spans over two decades at cutting edge of AI, robotics and advanced computing with, with past leadership roles at Johns Hopkins apl, uh, Accenture Federal Services, uh, where she helped shape the AI strategy for nearly, uh, every cabinet level agency. Um, she holds 10 patents and uh, and a PhD in physical chemistry. So needless to say, we've got a lot of ground to cover. Uh, Jennifer, it's an honor to have you on the podcast.

Speaker B: Thank you. It's an honor to be here.

Speaker A: Awesome. Now, before we unpack your journey and your role@empower AI, uh, I like to always start with something pretty deep. Uh, what did Jennifer Sample have for breakfast this morning?

Speaker B: Nothing. I have not eaten anything yet. Oh no. What's the plan for after this?

Speaker A: Is that par for the course or is this. Today's been a hectic one?

Speaker B: Today was just a very busy day.

Speaker A: As soon as possible, I had a, uh, it was a chocolate espresso filled croissant. Uh, yeah, which was, which was dangerously delicious. I shared it with my two year old. Uh, but my wife, uh, we're big on going to like some farmers markets and, and kind of, you know, touching base with local bakeries and the, this was one that just kind of jumped off the shelf. I'm a sucker for like a sweet, sweet breakfast. So, uh, this one, this one, uh, did uh, the trick. Um, but um, yeah, we'll wrap this as soon as we can so you can get to that meal. Um, but uh, you know, before um, uh, we, we get into more about like what we're doing, what you're doing here with Empower AI, uh, in true how we hatch fashion, we, we like to rewind the clock a bit and we'll uh, ask you to take us back to, you know, where did you grow up and walk us through that early part of your journey. Uh, and what led you into the path of chemistry to AI and specifically now. Very curious. How is space involved in that transition?

Speaker B: Sure. Where to start? I grew up in Hanover, Pennsylvania, which is near Gettysburg. So the closest city, uh, was Baltimore. And I've always been interested in science from the very beginning. I was the first person in my family to go to college. Ah. And I think just the power of science to explain everyday phenomena just is what hooked me. Um, I can remember being in. I have so many early memories that are really positive with science. Like being in gifted, um, and talented class. And they, ah, poured water into a Dixie cup and kept pouring and pouring until it kind of, you know, as it does like try this with your daughter since you have a two year old. It kind of like, you know, uh, extended beyond the top of the cup but didn't come out interesting. Yeah. And we were just kids. I mean this was literally elementary school. And the teacher asked, the point of the exercise was brainstorming. She was just trying to teach us brainstorming. Right. This is, I'm dating myself. But it was kind of like, oh, give me your ideas. Why is this happening? You know. But the answer was surface tension. It's the surface tension, right? I mean, just things like that, uh, you know, just completely blown away by, by science's ability to explain things that I saw as a naturally kind of curious person. So that's what I pursued. I really liked chemistry in high school. Again, you know, I remember being in the classroom, kind of listening, kind of not hearing like screech on the whiteboard. And it was a very echoey room because we had kind of a classroom and then a, behind it and the lab was completely empty. And you know, it's just kind of, I just remember like the chalk and the echo. And then the teacher said, and this is how a battery works. And then suddenly it was this, you know, redoctoring moment. Yeah, exactly. So, so that was pretty natural for me to, you know, choose as a major in college and to study. And um, I always, you know, kind of my overarching philosophy is to make the most impact that I can in the world. And so, you know, for me I wanted to keep going. So I went straight through, uh, to the PhD. And that next big thing has always excited me. Not just science, but, you know, that science is going to change the world. And so when I was coming up through Undergrad and my PhD, it was nanotechnology, science of very small things. And so that's how I ended up, you know, doing a PhD in physical chemistry, uh, because that's that, um, space between physics and, uh, understanding the properties of very small things, but then also kind of what can you do with them? That definitely helped me see how small components can come together to shape complex behaviors and applications. That's the beginning, and I will get to the space part. I joined Johns Hopkins Applied Physics Lab out of my PhD specifically because I wanted to do something that had. I had the intellectual freedom to work on what I wanted, sort of in an academic style, but then also, you know, um, more application. Not so applied that I was working on antimicrobial trash bags, which was one of my other job offers, but applied enough to see that satisfaction of, uh, you know, the science turning into something. I started running experiments myself and in my very early days because there's so many tough problems out there that had never been solved. Bring in the nanotechnology person. Right. And one of those was NASA's Parker Solar Probe mission. I don't know if you're a space person, so you look into this. I think NASA since the 50s had wanted to launch a spacecraft to the sun. Uh, you know, obviously solar winds and solar flares, these shape life on Earth. But it's super hot sun and just you know, really very difficult to figure out a mission. And I think many different labs had had their turn at it. And by the time it came to John Saffield's Applied Physics lab does a ton of space missions. And so by the time we got a crack at it, it was an internal R and D program. The hypothesis was carbon nanotubes as a heat shield, uh, could help with this survivability issue with the radiation and all of that. Um, and you know, in 2018, I got to take my family down and watch it launch and keep track and it's. Yes. So that's that kind of full, that full cycle of, you know, idea, ah, exploration to application. And that kind of got me hooked.

Speaker A: That's great. Yeah, that's actually on, on my wife and I bucket list is to catch a, a launch in person. Um, so we'll have to ping you on some notes on the best place to be to experience it, but what a really great experience and also just kind of coming full circle with kind of seeing some of the work that you've applied, ah, in a production setting. Right. And then seeing ah, it on the big picture here with an actual launch. It's incredible. Um, and I'm a huge fan of Johns Hopkins, uh, Applied Physics Lab. We've had a lot of guests actually on the podcast Podcast previous, uh, alumni that have worked in, in this kind of a setting and then, you know, making transitions either into consulting or startup industry. Um, and so I'd love to just kind of pull on that there a little bit, um, before we, you know, talk a little bit more about empower AI, you know. So you were, you've been working on some really interesting projects during this time at Johns, uh, Hopkins apl. Um, some of that work was supporting uh, DARPA for research and AI, biomaterials, human machine interfaces. So in that setting, kind of how do you balance like the visionary side of emerging tech with uh, this reality of more mission driven outcomes?

Speaker B: That's a great question. You know, all the DARPA work is high risk, high reward, right? And so you make a lot of bets and they don't all work out. And I think part of the reason I found myself in leadership roles at APL was because of the big bet, you know, with the solar probe which had, which had worked out in the end, you know, the heat shield didn't involve carbon nanotubes. But all of the work that went into solving the material science for the heat shield, you know, was kind of like that. That's the whole point of this R and D is to eventually transition it into something and solve the problem. So it was super fun working with darpa. Uh, you know, you know about darpa, everyone does. They're wildly ambitious for sure. And so, you know, some of the work is still ongoing. We were part, you know, as my career kind of evolved, you know, from scientists to leadership, I branched out into other areas including AI, robotics and Autonomy Quantum, um, and worked with darpa, uh, on some of the. Right now all of our lives have been transformed by, or are being transformed by large language models. But before we had large language models, we had, you know, low resource languages and uh, you know, computer vision and things that are just kind of being subsumed in, you know, by AI right now. And so it was really fun to do a lot of that work and contribute to the general body of knowledge that advanced, um, those technologies. At the end of the day it is tough, right? It's really tough because sometimes you just. Something that doesn't work is actually a result, right? It's a finding. It tells you what not to focus on so that you can refocus on what you should focus on. And I think for me, after doing that for I think 19 years I was at APL, I just. You get tired of synthetic data, you get tired of toy problems and that huge impact doesn't come along that often. And I think that kind of combined with the fact that I was not really using my PhD at that point. I was leading a team and, you know, building a team and doing the kind of optimizing the organization for innovation and kind of a lot of business things. That's kind of what got me, you know, looking again, you know, how can I make the biggest impact from where I am right now? So that was kind of part, um, of my transition from APL to Accenture and eventually empower AI.

Speaker A: Yeah. What, you know, 18, almost 19 years in a place. I mean, it's a home at that point. Right? I mean, those co workers are family, uh, at that point. So a, ah, really great. You know, you don't hear about tenure like that these days. There's really a lot of folks that are just kind of, you know, a couple years here or there and then on to the next thing. Um, so clearly, you know, there were things that you were working on that really just, you know, you're extremely passionate about, and then getting that exposure to areas of, you know, innovation that a lot of folks wouldn't be able to have access to. And, uh, in. In this kind of like a lab setting. So really neat. But let's talk about that transition, right? Because you were kind of itching for something a little bit more. And so you make this move into Accenture, you know, big four consultancies. Um, you know, what were. What were some of the things that you got introduced to here? And what was it that was kind of scratching your itch in that environment?

Speaker B: Yeah, it was definitely a growth curve for me. Around that same time, I had gone, um, back to school actually, to learn the business side. So I joined Accenture, you know, straight out of, um, my executive MBA at mit. And that just really opened my eyes to all the sorts of impact that we can make besides, you know, the relatively narrow scope of applied R and D. So I was very much looking for that. I had been spending my entire career on revolutionizing technology, progress through technology, mission through technology, like we were just talking about. And it's hard, and a lot of it doesn't work out. But then what I saw in terms of kind of the gap, uh, between commercial and government technology just didn't really connect with what I was doing. We are solving really important research questions. But a lot of that gap, a lot of the impact that you can have in the federal space is just digital transformation. I know that's kind of a cliche word, but getting rid of Technical debt and improving workflows with technology, connecting data so that you can train up an algorithm, um, on it. And so that's very much what I was looking for, is that kind of bigger impact. I knew that I wanted to transform the government using AI and I didn't quite realize it wasn't AI research, it was actually, you know, AI, AI implementation. Figuring out how to responsibly adopt AI. What are the barriers? Why can't we do AI right now? Maybe it's. We don't have the data governance in place in order to share the data across the organization. Blending that kind of human side of. Why hasn't this happened yet? With the uh, what's possible from a technology perspective for me is very deeply, deeply satisfying because it's possible to make a even more impact that way if you can influence kind of the decision makers and the stakeholders that are setting some of these policies or making some of these decisions while informing them about the art of the possible from a technology perspective, maybe even introducing them to other clients that are further along. It's a powerful way to make an impact and it was deeply satisfying. They are doing things like, um, rewriting legacy mainframe from the 1960s code. Assembly. Yeah. Yes. So, uh, yeah, that. Very satisfying. Definitely a bigger impact. And um, just very, very different, but also really fun.

Speaker A: Yeah. And obviously consulting is a, uh, you know, a skill set that really rounds folks out as a, as professionals. Um, you know, there's, there's things that I've always seen that, uh, you know, startups, for example. Right. Will oftentimes look at bringing folks in that kind of come out of this consulting environment because oftentimes the skill sets that are really sought after are this kind of like techno functional skill sets. Right. So folks that can kind of, you know, really understand the technical problem at a deep level, but also be able to relay and communicate that to the customer. Um, that is really tricky to kind of flesh out with, you know, talent sometimes. And so, so, um, I really respect, you know, the individuals that kind of spend a decent amount of time in a consulting environment, especially really well known if it's like a Deloitte Accenture or something like, like that to, to, to give these folks this kind of a, um, uh, just a, a sharpening of, of that skill set to interact with customers at a deeper level and just being able to explain that. Right. I think that's really oftentimes a problem area that we see is like, yeah, you know, technical folks can understand it themselves, but to be able to translate that, in layman terms, is a whole nother skill set in itself. So consulting's really, uh, a, uh, neat platform to sharpen that skill. So let's talk a little bit about, you know, so you've gone from John Hopkins APL to Accenture. So what was the spark for this turning point that led you to join Empower AI and step into this CTO role?

Speaker B: Yeah, I mean, for me, I think it was a culmination of everything that had come before. I love that, uh, word techno functional because I had that. And it's not, you know, it's not as easy as it sounds, I guess. And so after years of, you know, building and leading and advising and translating across technology and kind of functional and operational, I don't know, boundaries or artificial boundaries, I kind of reached a point where I wanted to, um, kind of set the strategy, own the architecture, not just be part of it, but also the thinking, the talent, the priorities, kind of the whole thing. So, um, that's really what brought me here. I think that I saw it as an opportunity to really kind of honor both my technical and my kind of functional kind of people and business interests. And again, a way to make more impact. Right. I mean, uh, as a consultant, you can transform one aspect of one client's use of AI or use of technology, but leading the technology strategy of an entire federally focused company with many contracts, many different customers is a way to make more impact. So that's kind of.

Speaker A: Yeah, yeah, It's, It's a. You're doing it for your. It's almost like you're doing it for yourself, not for somebody else. Right. It's, it's ah, it's like you're building, you know, for, for your organization versus uh, implementing solutions for somebody else's organization. Um, and so that is, ah, uh, an empowering feeling. And, and, and obviously, um, you know, this is a, this is, you know, so a new kind of opportunity for you. Uh, can you just kind of give us a little bit more about, you know, for. Is Empower AI focused on today and kind of lead that into an overview of its mission and how you all are applying emerging technology to federal challenges?

Speaker B: Yeah, happy to. I mean, at the end of the day, we do AI enabled IT services and as much, uh, AI services as the, as the market can take. Right. I mean, that's. Things are a little different now because again, in the past it was all of this influencing and showing the art of the possible and coming up with a prototype that really resonated with that exact Pain point to get someone starting to think about it and now it's like, use it now. So that's been great. It's definitely good timing. I would say we're being forced to ask kind of sharper questions. Like when you think about kind of um, augmenting workflows with technology at scale, it's important to think about what will be the devil in the details as much as you can, you know, in advance. And will this really endure under pressure and to have that discipline, um, to be more thoughtful. I think for us it's really all about getting customers to the outcome that they're looking for. There's no shortage of tools. It's easy to look and see all the tools that are atoed and fedramped and there's amazing software products with amazing capabilities and functionalities. But our customers end up in this environment where they've got so many different systems and tools and some of it is outdated and needs to be updated, some of it is cutting edge, but all the features aren't turned on. Whatever the situation is, they're not necessarily getting that outcome. Like how can we know when the WI fi is about to go out for a vip? Mhm. I mean the answer there is it could be in power data, Maybe it's in ServiceNow data to just kind of take that problem, dig in, diagnose it, have the methodology for approaching things like that and then to have, we basically have a suite of what we call technology accelerators which are kind of like partially built. They're fully built tools, they're almost like partially built products. They're not um, intended to be software products. They're just intended to accelerate that time to value, to really give you a low risk, high impact way to try something. And so a great example we're calling it QueryGov, uh, is just an FAQ, large language model. There's infinite examples of where you need to reference a policy to do your job. And should you have to look through, like find where the policies are and then look through them all. Not anymore. Not that we've all, not now that we've all used ChatGPT. Right. All you need to do is just, you know, put an LLM on top of that repository and make sure that it's behaving appropriately and citing, you know, references or whatever the constraints may be. And that just, I mean it, you know, we're measuring the tremendous impact that it's had, having and it's not just saving time, it's improving morale because people are happy when they're uh, able to focus on those, you know, more strategic decision making tasks versus the repetitive ones. So you know, very good uh, predictive IT technology accelerators like that.

Speaker A: Yeah, I like the uh, reference to outcome based because you know, there are just this exhausting list of tools that are available to folks and sometimes what we've seen is like, it's counterproductive. You know, it's almost like, like decision fatigue almost. You know, you've got so many things that you're looking through. It's like, it's like walking down like the cereal aisle. Like it's, it's like there's almost too many options here to pick from. Um, and you know, if it's not being utilized the right way then, you know, this is why I think there's also this big push for the role of like customer success. I think it's just such a f. A fantastic role for product companies specifically because you know, the tool's only as good as the user is going to be like with the ability to know how to navigate it. And so um, it can be uh, just a counterproductive uh, innov innovative resource if it's not um, uh, you know, kind of introduced the right way and kind of whittled down to what's really needed to solve those problems. Um, and so approaching it from an outcome based uh, approach is really intriguing. Um, now obviously there's um, you know, there's, there's challenges with you know, modernizing technology and government. We talk to a lot on this podcast. Specifically, uh, if it's you know, autonomous systems for you know, uh, unmanned surface vessels, for the Navy to uh, counter drone technologies. It, it takes a lot of work to cut uh, through some of those different layers of bureaucracy. There's a lot of movement being made in real time to help better, you know, acquire software or uh, be able to introduce commercial technology into uh, some of these big picture problems across defense and national security. What are some of the biggest challenges or opportunities that you've seen in deploying AI across government?

Speaker B: I think, I mean, you know, I wouldn't be a proper CTO if I didn't say access to data. Honestly, you know, you kind of want to say the human side, but honestly, even more than that, that uh, it's access to data.

Speaker A: We can, one of the biggest data, uh, aggregators is the government. We've got so much data. How do we get our hands on it? But do it the right way?

Speaker B: Yeah, absolutely. And what are the policies and who are the decision makers? And does Some organizational change need to happen first until you can unlock that data. Um, because you can build all the accelerators and you can do all the proof of concepts and demos, but you really need the actual data that the customer is dealing with in order to really deliver that value and that outcome. And for whatever reason, I mean, just note to the audience, that is the long pole in the tent. That is the thing that takes the longest. So start early.

Speaker A: Cool. Yeah, that's great advice. I just want to kind of check some boxes off, uh, around, uh, Empower AI. So what's kind of some quick hits in terms of maybe like the company headcount, um, uh, where you all are kind of headquartered, based out of uh, some of your biggest customers. Stuff like that would be helpful.

Speaker B: Yeah, so we're about a thousand people. We are headquartered in Reston, Virginia. Our biggest customers, we do a lot of, as I mentioned, IT services management and focusing on outcomes and AI enabling that. So we do a lot of work with, with GSA right now. So it's been very interesting as they've been uh, experimenting with AI and changing dynamically, supporting them. We work with the Pentagon, disa, the um, help desk there, uh, jsp, help desk there. And those I would say are our biggest customers. And then in kind of the more LLMs and more. I don't want to say adventurous and I don't want to say R D. That's not quite right. But those early stage prototypes where you're kind of confining everything very tightly so that it works well, you can learn from it, build momentum and then go from there, that's so critical for AI adoption in that arena. We're working with, uh, dla.

Speaker A: Very cool. And you mentioned REST in Virginia, but I'd imagine there's folks across the country that are, um, working with you all.

Speaker B: Yeah, we have people in Fort Huachuca. We do some work with the army out there. And GSA has offices all over the, all over the country. So. And then you know, some of our customers up in Fort Meade and Maryland. So. Yeah, all over.

Speaker A: Very nice. Um, okay, so I guess in, in kind of a, a little bit of a tidy up before we, we, we close with some, some fun kind of rapid fire Q and A. Um, I, I wanted to just, you know, kind of looking back across your career, you know, from you know, you got patents to programs to people, you know, what advice would you give to the future tech leaders that are navigating innovation and some of these high stakes government environments?

Speaker B: I would say don't Underestimate the relationship piece. I think as a technology person, we go straight to, you know, what is the system, what is the piece of software, you know, what can we do with the data? But the people are absolutely critical. You can't really achieve any of those things without people being able to communicate the impact, being able to understand their perspectives. I would say to really get things done, it's important to be able to understand culture and people and technology. So just keep that in mind.

Speaker A: Yeah, I can't, I can't, um, you know, kind of chime in on that as well. Just the, the relationships piece, um, especially. And this is, you know, an area that I hope, uh, you know, does come back to, to fruition at some point. Is that, that in office environment, that collaboration in person, um, I think for the, the younger generation that are just getting their, their career started, right, you can think back to how valuable those, you know, the happy hour or just, you know, the going out to lunch as a team, and those relationships that you build with those individuals and how they help navigate or help craft, you know, your career path down the line is just, um, can't. Can't say it enough. So hopefully those, um, yeah, those are, you know, coming back into office is a thing and we're seeing it quite a bit. But, um, the idea of hybrid also is also very appealing too. So which you mentioned, that's kind of part of the, that, you know, some of the culture that you've got is a little bit of this hybrid style environment. Um, cool. All right, well, I'm gonna put, uh, a bow on it at that point. I want to transition this into this, you know, final segment. Uh, it's, uh, a, It's a fun segment. We call it the five second Scramble. Uh, so you basically, you know, try to give us your answers within five seconds. Quick m. Hitters and some are going to be business, some, Some personal. We're not gonna get. We don't get too personal on this podcast. So, um, let's kick it off with, um, the first question. Uh, so if Empower AI were an animal, uh, what would it be?

Speaker B: It would be, uh, I don't know. A rhino is what's coming to mind, but I can.

Speaker A: A rhino is, has a, is a presence, a massive animal. So I like that. I don't know why it came to your head, but that's great. Well, we're going with rhin. Um, what's your favorite part about the culture at, uh, Empower AI?

Speaker B: Definitely the people. Yeah, the people are amazing here.

Speaker A: What Kind of technologist thrives at, uh, empower AI.

Speaker B: Someone who can think out of the box, look at things that others have seen, but think what uh, no one has thought before. Someone that loves challenges.

Speaker A: Very cool. What kind of uh, tech roles are you hiring for over the next six to 12 months?

Speaker B: AIML engineers, uh, solution architects. Those are the big categories. Uh, some developers and probably some cloud engineers and cloud architects soon.

Speaker A: Very cool. Uh, what is the first app that you check in the morning?

Speaker B: I usually just check like my calendar just to see what I, I know, I don't, I don't get up and

Speaker A: get on social media like a true business person. I do the same calendar and then mail. Um, what's a book or you know, business kind of podcast, uh, that you always recommend beyond the prompt.

Speaker B: Uh, if you haven't, uh, it's, it's a podcast about how to transform your business using AI and it's just people that have literally done that either out of necessity because they were a startup and they built, you know, agents that could do their marketing or you know, um, it from a philosophical perspective. Like there's a quite a variety, you know, there's some established adoption stories and then there's some kind of like early ones. But I, that's my favorite, I would say.

Speaker A: Cool. Yeah, I'll check that out. Um, what, uh, is the worst fashion trend that you've ever followed?

Speaker B: Worst fashion trend. Oh my goodness, that is a tough one. Um, I, I'm a runner and an athlete and I, I, I think it's safe to say I'm not a, uh, big fan of running skirts. Although I have worn them in the past. I would say. Like running skirts.

Speaker A: Yeah, running skirts. Good to go. What's your go to coffee order?

Speaker B: It's tea. Green tea. Tea. Yeah.

Speaker A: Nice. Yeah, nice switch. What, um, is a unexpected, uh, hobby that you've picked up recently?

Speaker B: That's a tough one because I started a new job recently. But I will tell you, here's a good one. Um, the airline miles game. Like people get really into this and I always was just like, but it's very fun. Um, if you get into it and you research like you know, nerd wallet or wherever you look and get your information, you can, you can get, get first class flights for like, it's incredible.

Speaker A: I recommend, yeah, it's a good hack. Uh, I'm a, uh, big, big into the like the American Express card for the, the sky miles. Last, last question and, and maybe this has already been answered throughout this conversation, but what was your dream job as a kid.

Speaker B: Oh, this is my dream job.

Speaker A: You're living it. Awesome. All right, well, that's a. That's a good closing note. Uh, Jennifer, I wanted to thank you for spending the time with us today and kind of talking through your journey. Um, I think it's a. It's a powerful example of, you know, this mission aligned innovation, and you kind of, uh, are a great representation of that. So we're rooting for you, rooting for the team at Empower AI and can't wait to see, uh, what you all do next. Thanks for joining us on the podcast part.

Speaker B: Yeah, thank you so much. It's been a pleasure.