Scott Kupor & Arun Gupta on Tech Force: Rebuilding Government Talent | American AI Festival 2026

Thank you. Thank you guys for having us, by the way. We're really happy to be here. And, you know, it's always fun to be with a good friend on stage here. So maybe here's, I think, where we should start, which is, look, we're here at an AI conference. I don't think either you or I are going to break any news on some new cool AI technology today, as much as we might think so. So Arun Gupta, we both come from the Venture Capital Committee. So like we play, play technologists, but we're not there. There's many people in this room who are way smarter than us. But the reason I think why we're here and what I think we should talk about is, as we know, for the technology to progress and to ultimately get adopted, fundamentally the question is, do organizations have the right talent to be able to develop applications that are AI-centric, to be able to actually modernize everything we do? And you and I obviously have a particular predilection for the government, helping the government achieve achieve these objectives through the use of talent. So with that framing, maybe I think maybe the best way for us to start is maybe just to give some perspective to the audience on what is Noble Reach like? What are you guys trying to do? And then obviously we can kind of dovetail that with how we are partnering together with respect to kind of some broader federal government opportunities on talent.

Well, thank you, Scott. And, you know, I think Scott and I share the fact that, you know, having a venture background, the one thing you learn is that it's less about the idea, less about the market, but it's always about the team. And when it's about the team, it's about the talent. And so that applies not only in venture, but I think it applies in all organizations. And the quality of talent matters. And, you know, especially in the case as we think about a tech-forward government, the quality of talent matters. We can talk technology, we can talk policy as much as you want, but if you don't have the right talent in the room, none of that matters. And I think both Scott and I would say that we're surprised at how much— how little focus sometimes is placed on the talent piece, when in our mind that's probably the most important piece of the equation. Real quick on Noble Reach and in the lens that I kind of come at this from. My dad was in public service, was in the Navy for 40-plus years, so it's been in my family. I was a VC, like Scott, for 25. And saw the magic of the entrepreneurial mindset and bringing this together. Then I was teaching at Georgetown and Stanford and saw the academics' language that they were using, which was still kind of setting the table for a binary choice for folks, public service or private sector, doing something where you make money or you give back and serve. You know, I think into what we're trying to do at Noble Reach is kind of break that stereotype and to show that folks can do both. And that frankly, if we're going to have a tech-forward government, we need to create pathways to enable both to happen. And students need to feel that that's an opportunity so that you can start your career in public service, and that could be additive to opening up opportunities in the private sector down the road. And if you decide to go into the private sector, that that's celebrated. And then if you stay in public service, that's celebrated. But it's the only way that we're going to really bridge the cultural gap and the trust gap that exists between the two. And ironically, it's the superpower that we have in this country. Like, we create talent better than anywhere else in the world, and we innovate better than anywhere else in the world. But what we haven't done is connected that to government. And this is why I think the work that Scott's doing is really important, so that we can solve problems at scale. So that our top talent feel that they can go in, but they're not having to go in forever. This is the key piece. When I think what I noticed when I was teaching is students are hearing that government is selling careers, but what students are buying are experiences, 2 to 3-year experiences. So at Noble Reach, what we are doing today, we have a half a billion dollar foundation. It really started with that notion of like, how do we change that mindset from why to wow? Like, why are you going into government? To, wow, you got selected to go to government, right? And it, it being around, having private sector partners around you that send the demand signal that we value programs like this, that you have agencies that we place them in with strong mentors, skill development exercise. So we've now run two cohorts. We've got the privilege— and was chatting with Scott as he was taking on this role of looking at— I feel like we're in a moment in time to scale this. And so TechForce couldn't have come at a better time. And I think when we look back on it, not to be hyperbolic, I think it will be one of the most significant changes in mindset as to how we engage public and private citizens as far as how we allocate talent in our country.

So let me frame it— that's super helpful. Let me frame it from the federal government's perspective. Perspective, just so you have a sense. And people may know this or not. But so if you think about the federal government, we have two— we have a lot of challenges, but we have two very fundamental challenges with respect to this particular issue. One is we have a major demographic problem in the federal government. So if you look at the under-30 age cohort in government, there are about 7% of federal employees who are under the age of 30. For any of you who are in the, you know, non-government sector, if you're in the private sector or particularly if you're in the startup community, I can tell you in the broad U.S. federal workforce, government— sorry, non-government, but just general workforce, that number of under-30 population is about 22%, 23%. So if you just look at where we are in government today, by a factor of 3 to 1, we are massively under-indexed on early career talent in government. On the other side, again, if you guys— for those of you who may not be familiar, if you look at the other age demographic, which is people who are, let's call it, over the age of 50 within 5, 10, 15 years of retirement, close to half of the federal of the workforce is in that bucket, basically. Again, if you look at the US workforce writ large, that's about 33%. So again, we are way over-indexed in terms of kind of people who are pending retirement in the next 5, 10, 15 years. And we're just literally barely keeping pace. So thing number one, as we think about this, and obviously Arun and what Noble Reach are doing, is we have got to find a way to figure out how to get early career people to decide that, government is actually something that is exciting, additive, challenging, career developing. And I 100% agree with what Arun said, which is the story that at least I believe we have been selling in federal careers has been a career. Come to the federal government and you have a 40-year career. I, you know, Arun and I, you know, are completely simpatico on this. Also, we both have kids in this age demographic. Like, I don't think anybody in this early career demographic thinks about a 40-year career when they're trying to find a job, right? People are looking about like, How do I learn? How do I develop? How do I meet new people? How do I do things? And so a huge part of what Noble Reach was doing pre-us getting here, and obviously what we're trying to now augment with TechForce, which we'll tell you about, is exactly that, which is demonstrate to this population of smart people that come to government for whatever period of time is appropriate for your own career development. And if you decide after that you want to go back to the private sector, we should celebrate that. We should help you get a job in the private sector, and we should demonstrate that the skills you develop in the federal government are completely applicable and, you know, kind of desired by the private sector. So that's kind of thing number one that from the federal government perspective we're trying to solve. The second challenge we have, which is obviously specific to this conference and this topic, is we have a huge dearth of what I would call modern software engineering skills, modern AI understanding, data scientists, you know, all the things that, again, for those of you who are in the broader tech ecosystem or startup community, you know, just go look at your engineering organizations and the skills you have there. We have just— we are under-resourced in the government in those areas. And as a result, as many of you probably know, because we have traditionally not been able to attract and retain those individuals, we have a massive contractor and outsourcing kind of problem in government. So just to give you a quick sense of numbers, we spend about $250 billion a year on roughly 2 million employees for the federal government. We spend $750 billion a year on contractors. Obviously defense is a big portion of that, but the other huge portion of that is IT broadly. And so essentially we have decided that we cannot find and get modern software and modern technology talent in government, and so as a result we're willing to basically just outsource that to other companies. And some of my best friends are outsourcers, so I have no— that's a fine thing from the outsourcer perspective, but I certainly believe that if we could actually inculcate that type of talent in government, We would all be better off. It would be, quite frankly, much more efficient from a cost perspective. And retaining that knowledge inside the federal government would be helpful. So anyways, just— that's just a little bit of framing of the problem that I think, you know, from each of our perspectives, we're trying to solve.

I'd like to pick up on a stat that Scott just laid out, which is, you know, less than 6 to 7% of tech workers in government are under the age of 30. When we started doing what we were doing, you know, I think people looked at that data and came to the conclusion that, oh, that's because this generation doesn't care. Right? And we fundamentally had a different view, and I know Scott shares that view, which is actually I think this generation cares and arguably cares more than previous generations. It's just the infrastructure to connect them to these opportunities hasn't been modernized. Right? And so that was really the challenge that we were trying to go after. And again, we were all looking at the same data. People are just coming with different conclusions. So many times people would ask us, how are you going to convince these kids to want to go to government? And I'm like, you're actually asking us the wrong question. This is before Scott got in his role. Our harder problem isn't getting kids excited. It's about getting placements.

Right?

Our bottleneck was on the other side. And so this is why what TechForce is doing is really opened up the opportunity set quite a bit in amplifying this. Because now you can take the— you know, the irony is we had demand on the student side, we had demand on the agency side. But what was broken was, was the connection piece, right? Like, and that includes Pathways, includes the language you use. And I think that's where we're, we're now systemically changing and I think has great opportunities here.

Yeah.

But let me ask you, Scott, I mean, maybe for folks, Here you are, VC on the West Coast. You know, this is not a light commitment, right? To decide to kind of come in and take this role. As you kind of come in, what one inspired you to want to do this? And, you know, what are you hoping to be the enduring change?

Yeah, I mean, look, it's very simple in my mind, and it's a lot of the things we've talked about, which is, I am, look, I've got a, I'm concerned generally about obviously how much money we spend in government and I, you know, everything. Look, my reason to come here is I believe that we can actually get the right talent here, we can use technology in an appropriate way, not as a substitute for all human labor, but to actually augment what we're doing so that we can deliver actually better services to the American people without actually bankrupting our kids and grandkids. Like that is my fundamental, like driving desire for why I'm here. And my second one is exactly what you and I have talked about, which is I would like to convince every, you know, kind of person who is early in their career that you should come to government because you can tackle like the world's biggest problems, the most complex issues. You can be surrounded by smart people. You can grow and develop your skills. And then you actually create option value for whatever you want to do. If you want to stay in government, that's fantastic. Let's find you a job in government. If you decide you want to go to the private sector, we should do that as well. Like, that service ethos is, to me, is just incredibly important. And I think we've just lost that. And again, as you mentioned, right, look, you've typed into a cohort of people who are interested in that, and yet, and we haven't connected on the federal side. But I think we've got to go tell that story more broadly. The other thing we're doing, as you know, at OPM, and so for those of you guys who don't know OPM, basically think of us as like we're the HR organization for the federal government. So we're responsible for promulgating policies around anything talent-related for the roughly 2 million civilian workforce in the government. And then our second job is getting the right talent in the right seats, basically, right? Of which, obviously, these programs are important. The other thing that we have to do— to me, there's kind of two problems that we're trying to solve at OPM. One is we've got to go market to and tell the message to early career folks that we've been talking about. And we keep mentioning TechForce. We're going to, at some point, tell you what TechForce is. But then the second thing we have to do is we have to make sure the environment they come into in the federal government is responsive to to the type of environment they want, right? So we are making changes to everything in government from how do we hire people. So I'll just give you an anecdote. As you may know, traditionally in government, we have hired people based on their own self-attestation of their skills. So historically, if you applied for a software engineering job in government, you would fill out this form. And you would say, yes, I know Perl. And I know C++. And we would say, fantastic. That's great. You're hired, basically. What we are now doing, and TechForce actually is the first literally example of this, is just like you would in any other normal company, when you apply for a software engineering job, you take an assessment. We use a third-party commercial tool to say, OK, like, show me that you can actually solve some basic computer science problems and stuff like that. So everything on the front end of merit hiring to then what does the performance management system look like. So again, a huge problem we have, which I think really discourages early career people, is every job in government in terms of promotions and in terms of leveling is based upon college degree requirements and minimum tenure requirements. So the canonical problem that we have, which the private sector doesn't have, is for you, you know, if you all find some fantastic 19-year-old or 20-year-old who dropped out of college who is like the world's, you know, greatest gift to programming, you hire that person and you pay that person what their market rate is. In the federal government, we would pay that person $40,000 because they don't have a college degree and they have no years' worth of work experience, both of which in my mind are like false proxies for merit. They're just— those are shorthands. They're not actual demonstrations of merit. So one of the things that we're doing, again, which I think will make this more attractive to early career people is we are eliminating all those degree requirements and tenure requirements. So if you are a fantastic engineer— and we're doing this, by the way, in TechForce— we're able to hire people at what's basically a GS-14 level, which is the the second highest, you know, the penultimate highest level essentially in government because we have basically said like we don't care about those things. We actually care about demonstrated skills and demonstrated merit. And so all these things, there's kind of two sides of the coin here, which is like we got to do top of funnel. And this is where, you know, all the work that Arun Gupta is doing and helping us is great. And we also have to solve the operating environment they come into to make sure that it's a place where when they do good work, they get paid, they get recognized, they get promoted, they get more experience. And that is what creates optionality for them. Do you want to comment on that, or do you want to tell people about TechForce?

Well, I feel like you should be the one telling them about TechForce, but let me give you my perspective of why TechForce has been helpful for what we're doing at Noble Reach and why they've been a good partner. But then I want Scott to get into more detail. So again, we're going and recruiting on college campuses. We're recruiting— from students. First year we had 250 applicants for 20 spots. Next year we had about 1,300 for 30. But this year now, you know, as you can see, like the applicant pool was way outpacing what we could find placements for. What TechForce has been able to do for us is being able to have one agency to go talk to, which then allows us to find placements across all the federal government. Whereas before we were going agency to agency. And in doing that, we've been able to create a force multiplier whereby we can now have a pathway to really excite folks to be able to come in. The second is with the platform of what they've created is pulling together an ecosystem. And I think this is a really important piece, is an ecosystem of private sector partners. And those private sector partners send an important demand signal to the students. Because they feel seen when they do this program. When they go through the program, they know the top firms that they're used to getting on college campuses value the program, right? They're offering credentialing and skill development exercises while they're doing the program so that when they leave, they can decide that they're leaving it with feeling like it would be a career enhancer for them, as opposed to feeling like someone's going to be asking, like, why did you go into government? They're going to be saying, like, why you got selected for TechForce, or you got selected to be a Noble Reach Scholar with TechForce. And that's a great experience, and you know that experience comes with things like mentorship, it comes with skill development, it comes with community, and it says something about the individual. Um, and so from our perspective, what, what's happening at TechForce, um, really becomes a larger platform to start to systemically change how we talk about early career going into the federal government. Now, for the more specifics, maybe Scott, you should talk about—

Yeah, let me tell you.

There's a lot more in the background. I've given you the high level.

Yeah, let me tell you what it is, and I'm going to actually have an ask for this audience, okay? So specifically what TechForce is, is we are trying to bring 1,000 early career engineers, data scientists, AI specialists. We will soon launch a cyber Rec. We're soon going to launch a design rec. But basically, the simple— when people ask me, what do we need? The answer is, look, we need everything that you all have in your companies, basically. Like all the skills that you have in your engineering departments, product managers, right? That's all the kind of stuff we need. So the goal is to try to bring 1,000 people in. We're hiring them in a 2-year program. And specific to Arun Gupta's point, the 2-year program in our mind was really critical, which is we want to demonstrate that the skills they develop here are generalizable outside government if they decide to do so. So we're not asking them to make a 40-year commitment here. We partnered with close to now 40 private tech companies, and we are very happy to have more partners. So if any of you are interested, please do let us know. What we're asking the private sector companies to help us do is exactly kind of the high-level stuff Arun described, which is help us with certifications and trainings, help us with career development for these individuals, you know, help us identify kind of how do we make this like an interesting cohort. So, you know, we'll have— and, you know, one month we'll have Satya Nadella come do a chalk talk to this team. The next month, hopefully, Sam Altman will come and do that. But things that demonstrate to this group that they are valued by the private sector and that create, you know, kind of learning opportunities. And then importantly, at the end of the two years, we are going to run a comprehensive job fair with all these partners. And if hopefully we've done our job right on the front end of selecting these individuals and then in the two years in government, They should be. My hope is that every single tech company is salivating over these individuals to hire them. And every venture capital firm wants to fund them to do their startups or whatever they want to do. That's the flywheel that we're trying to develop here. So the ask, so specific ask is, one of the things I've learned in government is there's no such thing as marketing budgets. You may know this. But it turns out, by the way, funny story, we wanted to get hats and shirts for TechForce. Because look, as a traditional Silicon Valley guy, to me, if you don't have a hat, you don't actually have a business. Business. Literally, like, I am not permitted to do that stuff. So luckily, I have the ability to come out of pocket and buy hats for people. And I've asked my friends, like Arun Gupta, to buy some hats as well. So what that means is, look, we would love all of your help. So if this resonates at all with you, go to my Twitter is @scooper. Go to @USTechForce. Go to @USOPM. I don't even know what Noble Reach is I apologize. But like, help us get the word out, because literally this is just a grassroots effort. We are trying to get everybody and anybody to help us kind of get the message out there. And then secondarily, if anybody's interested in, you know, kind of help— if you know talent that's interested, please tell them about it. If you have an interest in partnering with us, again, the goal here we think is pretty noble, and, you know, no pun intended, and we think you all can help us. Maybe what we should do, since I know we've got 5 minutes, If— let me see, are there questions, comments? I'd love to give people an opportunity quickly if we can. If not, Arun Gupta and I can obviously, you know, kill the time here. Yes, go ahead.

Go for it.

So,