In this podcast episode, Nand, a former founder and now a representative of the intelligence community, shares valuable insights on the intersection of the private and public sectors in the realm of national security. He emphasizes the need for better communication between startups and intelligence agencies, highlighting the challenge of understanding the specific needs of the intelligence community due to its inherently secretive nature. Nand calls for increased transparency, expressing the intention to broadcast the agency’s needs more effectively. He also addresses the integration problem faced by the defense industry, stressing the importance of modular software and systems that can seamlessly interface with existing infrastructure.
The discussion touches on the complexity of modern warfare and the changing nature of intelligence operations. Nand encourages startups to focus on enabling speed and scalability for commanders and senior leaders, rather than attempting to eliminate humans from the equation. He emphasizes the importance of viewing human capital as valuable and the potential for technological advancements to enhance decision-making processes. Throughout the conversation, Nand demonstrates a deep understanding of the challenges faced by both the private sector and the intelligence community, striving for stronger partnerships and shared goals in ensuring national security.
In this episode, Lauren, Hondo and Nand discuss:
- Intelligence community challenges
- Private sector integration
- Communication gaps
- Complex government infrastructure
- Modular software development
- Speed and scalability in warfare
- Strengthening partnerships
Nand Mulchandani is currently serving as the Chief Technology Officer for the Central Intelligence Agency (CIA). Mr. Mulchandani most recently served as the CTO and Acting Director of DoD’s Joint Artificial Intelligence Center. He also co-founded and was CEO of several successful startups– Oblix (acquired by Oracle), Determina (acquired by VMWare), OpenDNS (acquired by Cisco), and ScaleXtreme (acquired by Citrix). He has a degree in Computer Science and Math from Cornell, a Master of Science degree in Management from Stanford, and a Master in Public Administration degree from Harvard.
Business executives for national security welcomes you to building the base here thought leaders and practitioners discuss how we can ensure our shared security and prosperity. They’re shaping the future of the national security industrial base. Your hosts or Silicon Valley defense expert, Lauren Bedula, along with Ben’s Distinguished Fellow and former head of acquisition for the Navy, Marines and special operators, Hondo Geurts.
Welcome back to building the base learn Bedula, Hondo Geurts. Here again from Boston. MIT’s campus. I mentioned we’re here for a Harvard MIT conference. We’ve got a Harvard alum with us today Nand Mulchandani with us who is CIA’s Chief Technology Officer. So I think just the fact that you’re here shows how the intelligence community and DOD are trying to work together on these issues, which is important. Not has seen the issues we talked about from many angles. He was a tech executives spend a lot of time Silicon Valley, then got recruited by DOD to be the joint artificial intelligence centers Deputy Director than acting director for
CTO. Yes, CTO, CTO.
Yes, yes. And now over at CIA so Nand thanks so much for joining us.
Thank you, Lauren. Thanks for having me. This is a long time in the making. And I’m finally that we’re pulling this off.
It’s awesome to have you here with us, as we’re out looking over the Charles River. But even more interestingly, have you know, hundreds of folks, students, DOD, intelligence community, folks, all kinds of thinking on these issues. But my question to start off is a little bit of how does somebody wind up being Q? Where you the kid like always rebuilding your parents refrigerator or something? How did you get? Did you get to this? And was it even something you had ever thought was in the realm of possibility?
No, great question. Well, first, first thing off, I’ll say is I actually don’t consider myself to be Q. I think that title would probably go to the Director for Science and Technology, who I think I’m not going to name him or out to him. But he’s, he’s the current Q, obviously.
So you’re R?
Q star is something definitely not Q. Although it is kind of, you know, I’m in the Q like area inthe alphabet. Yeah, I mean, it’s one of those things where you just can’t plan your life to that level of detail and, and, and precision. I knew. So I started my career back in Silicon Valley, back in 1991, I worked at Sun Microsystems, have a computer science degree undergrad, worked in chip design work in compiler design, and always knew that I was going to do startups, my dad did startups, he was in the semiconductor industry. So I was predisposed to the tech space. I also knew that I wanted to be in government in some form at some time in my life. But you know, being a founder and CEO of multiple tech companies, you’re executing two inches away from your nose is the product working as the investor base competition getting the pink getting payroll done in two weeks, so you’re never thinking big picture big, big things, etc. And my path to you know, I did four startups, back to back and it was an intense 25 years didn’t literally lift my head up. And then the fourth one, when we sold it to us to Citrix I finally stepped back and I was like, well time to have have a midlife crisis. And like what is my like, unfulfilled life dreams that I hadn’t really worked through. And working in government and public service was always something that that I had in mind. So I went and quickly executed, sort of using getting two master’s degrees back to back as a way of pivoting, and so a year at Stanford at the business school, got a master’s there. And then the Harvard Kennedy School. What was exciting was the business school side of it was interesting, because it gave me a lot of formal structures and thinking around stuff that I had basically been like, basically ghetto engineering for 25 years.
Informed your scar tissue.
Yeah, exactly. Exactly. Like, oh my god, I finally learned accounting properly, or corporate finance. These are all cool things. I love the theory of kind of the practice behind it. The Kennedy School was awesome because it took me completely outside of my comfort zone. As a tech person steeped in tech doing startups. The idea of sitting through, you know, Graham Allison’s class or professor Ash Carter’s class or Eric Rosenbach, or Nick Burns, who’s our ambassador to China. Now, these are like incredible professors teaching about geopolitics and others things. And for me, this intersection between policymaking in the government and world events happening in this entire pivot that we as a nation are going through the US government is going through. And this collision between tech is the battlefield, and then policymaking government and geopolitics to me, it’s just absolutely super exciting. But again, a million years never thought I’d end up at the Pentagon with general Shanahan at JAIC. And, and nobody expects a call out of the blue from the CIA. Typically. Well, this is a good call, and surprising one. But Director Burns and the senior leadership team at CIA really wanted to go, I think bold, not that I want to call myself a bold choice, the right choice at the right time. But but they really wanted to go unconventional. The other thing, you know, so, so really stepping back, right, I mean, Director Burns, when he first came on board, sat down and analyzed and thought about and him along with the senior leadership team, really dispassionately like what are we doing as an Agency? What does the next 75 years of like, how do we think about this, and the conclusion was quite simple, which, you know, everybody sort of knows. But the simple strategies are the ones that really are the best ones, right? There’s a giant pivot from CT, counterterrorism to great power competition. And the fulcrum of that pivot is technology. And it’s only that technology is a multi trillion dollar industry that has powered productivity and world change for the past 20 years, it was an obvious piece. But this is where that intersection of now that China GPC side of it versus and the tech side were the two big ones. So this resulted in the creation of the China Mission Center, which was the first thing stood up the second was T2MC, which is the Technology and Transnational Mission Center, which got set up. And then the CTO position to support some of this pivot work was my role. And so Director Burns basically was Check, check, check. So now that we’ve organizationally, set up all the right, pieces, chess pieces, now the game is executing this pivot. So we’re at that we’re in that stage of the storyline. I’ve been on board, nine to 10 months, T2MC has been stood out about a year, and then the China Mission Center. So we’re early, you know, we’re we’re in the story book in the storyline. But we’re a year in Storyline now.
And I think that the reorginization was almost a signal to industry to be a bit more approachable, or edits to the importance of high tech sector and the like, can you talk a little bit Nand about your priorities as CTO?
Yeah, and I’d say, Lauren, that’s the exact point that signaling is so important. You know, the fact that, and we don’t think of it internally as a reorg, in as much as this is additive, right. So we added mission centers. So yes, it’s taking on a broader role. But it is truly the point. It’s a signaling function as well. The signaling sort of how how’s it going and what my priorities are, so we’re a super small team at the CTO function. So the C level functions which report into the Director, so I have a direct report to Director Burns, in that we serve everybody. Right. So we’ve got five directorates and 11 mission centers, my job and before even I got here, it’d be surprising to everyone to know that the CIA does tech at scale, and has done it for 75 years. Go to the museum. There’s incredible tech, there are teams are executing at depth and scale across the world and global 24/7 environment. So my coming on board is not to come in and grab the steering wheel from everybody. Everybody’s doing great work. But agency work is incredibly operational. We’re an operational team. So what happens though, in an organization like that, because culturally, when you’re so operationally focused, you’re executing two inches, five inches, 10 inches away from your nose, you do need a second set of people to come in and think through the changing world. What does the world look like two years from now, five years from now, 10 years from now. And given the size and scale of tech and the importance of it, you have to have players on the field that are playing different roles and different pieces but complementing each other. So we’ve been trying to come up with analogies to CTO function about like what we do and how we do it. So there’s this famous commercial, which is the BASF commercial, right? It’s, we don’t make the things you buy, we make the things you buy better. That’s the CTO role. I don’t own anything, I don’t run anything, I don’t operate anything. But I do feel responsibility for everything. And so what I’ve been doing is with each directorate with each mission center, the other side of it is because of the small scale of our team. And you know, we’re highly we’re, we’re a super, super small resource, we have the concept that I’ve taught our team to work through. And what I want us to work on is all the stuff we learn about in finishing school in the valley, right, and tech companies, the core of it is scale and leverage, right? So if we’re not executing, in the projects that we take on that generate 10x, leverage 100x leverage 1,000x leverage 10,000x leverage, we could be the cat chasing the laser pointer all day long. The beautiful part about the the Agency by the way, the stuff that I get to see and do and the people I interact with, I could spend 10 years there and just have a great time, crawling through every nook and cranny of every cool project. And see new tech and new stuff. I mean, it is mind blowing stuff that we do. And it’s a for a tech person like me, it’s like a kid in a candy store. But I’m getting paid by the US government to work not to enjoy my life. So the question is, the projects that we take on are sort of very high leverage points. So So what are those projects and, and some of that learning, I think, or at least point of view that I brought to the table was related to my experience at the DOD at the JAIC. So there was this concept that we need to bring this new cool technology called AI in right and it’s all there all the rage, everyone is talking about it so exciting and cool. So I used to have the slide that I sort of took us to show to the team, which is it was a it was the Empire State Building, which is my favorite buildings in New York. And at the top of it, I have a little curly brace that said, this is where my family thinks this is where my friends and family think I work which is the cool penthouse with the view. And at the bottom, it was like this is where I actually work, which is in the sewer and plumbing work, basically, put on a hard hat, get a plunger or a pickaxe and just start working with salt mines. What ends up happening is very often people want to chase after the shiny thing. But forget the fact that if we don’t have the right infrastructure, we don’t have the right plumbing, we don’t have the right foundation, the shiny stuff, can get it done. But it’s going to be done in a very, very poor way, it’s going to be done very not cost effectively. It’s not scalable, it’s not going to stick. So unfortunately, a lot of the work is at this garbage, you know, salt line layer, and we’re bringing that level of thinking, to, you know, with the teams who are ready and right, because we’ve got data strategy, AI strategies, we’re executing those, we’re starting to pull those together in a fundamental way, and changing the arc of how we build, deploy and run software. And again, being a software person, you can bring leverage and scale to software in a way that it’s very hard to bend hardware on. And we do a ton of hardware as well on that front. And I’ll sort of close the thought out. So software is elastic, and you can bring that leverage of scale, we can squeeze that lemon pretty hard and get those those things if you get the right architecture, the right setup and the right commercial partners, VCs and startups and everyone. And that’s a whole nother discussion to have. On the hardware side what’s awesome about being in tech right now is we’re seeing a revolution in hardware. Right? It’s the so I have this talk that I gave about kind of thinking about the research versus development, R versus D. And the other side of it is the membrane is what I call the hardware software divide. Once a piece of hardware or something that’s deep tech gets effectively digitized. It gets on the software, commoditization curve. And then you can apply all of the amazing software scaling techniques to squeeze that down. So biotech right now. It’s software, right other than the front end of the process, where you’re moving test tubes around titrating chemicals and stuff. With CRISPR and other pieces. You’re doing modeling and software. So now you’re applying software engineering techniques to biotech. Space, My God with the advances in materials science with the advances in CAD CAM and the tooling and other pieces there. We’re seeing a commoditization of space that’s basically turning it into a commodity. I mean, I’m seeing so many that space companies that it’s sort of like oh, yet another startup shooting stuff out in the space. There was a company we saw the other day, literally go to their website and the three of us can go book, a cargo like hold to be shot up in the space with whatever pay load we want, we could throw confetti, the three of us can just fill it up with confetti, drop it up into space and let that stuff fall off the earth. Like, that’s the level of game. So it’s an exciting time where even hard tech companies building Mach Five aircraft, like, you know, back in our days, Hondo and even Laurie, you know, like, we’re old fogies right . So, back in the old days, the idea of like private companies building a rocket in a spaceship and an airplane, or you get out of town, crazy talking. So I think it’s just an exciting time to be here. And our game of CIA, and I think any customer of our size and scale has to be being a smart customer. And understanding how to set up the right environment and conditions inside to be able to pull this stuff in. The the phrase that I love, funny phrase I’ll use for your audience here is there’s a VC, there’s a website called VC funded by life.com. You should all check it out. The tagline is “coupons for an urban lifestyle.” Basically, it’s 20 year olds, living off of free coupons for VC funded startups, free food, free drinks, free rides, free rent, whatever free quotes. The joke is like, we as a customer need to be like VC fund my life. There’s so much private capital out there building all this amazing stuff. Why should we be investing in certain things that we can buy from outside, I want five of them being developed, I want them fighting it out. I want the best one to win. And I want to be able to absorb that tech. So here we are.
Yeah, it’s uh, you know, we often talk about having to shift our mindset from exporter of tech to importer fast and really and point the one thing that I think happens in the valley, and you know, and in places where we’ve been able to operate at scale, and speed is creating your great network, and really operating in network speed. Can you talk to, you know, the Agency’s had a long history with In-Q-Tel and kind of being a pioneer and how to access a network that we had an access? I know, you guys are looking In-Q-Tel plus, you know, how do we get the rest of them? How are you approaching connecting to this giant network that’s seems to be expanding exponentially every day.
Hondo, that’s a fantastic point. And we’re thinking about a lot. And, you know, we’ve had a, I think, a 20 plus year relationship with In-Q-Tel, which has been phenomenal. And just like we were talking about in the other side, before was this idea of, you know, the venture capital community is huge, right? It’s hundreds of billions of dollars. And In-Q-Tel represents a very special partner to us, because we’re able to, well, first is this their staff, and that firm understands our business really well in a way that a typical Valley firm would never understand. The other thing they do is their situational awareness sort of radar is also focused on defense tech that traditional VCs may not see. So what as we’re thinking of this sort of new architectures and things going to your point and importer of tech, rather than thinking of ourselves as an exporter, we’re now starting to think about increasing the aperture with the VCs and startups. Now, it’s a systemic, there’s a hairball worth of problems that you probably know better than anybody else Hondo. You know, having, having dealt with this at scale, right, you’re dealing at a scale you dealt with. So the scale of it’s a systemic problem of how we do problem definition statements. Our level of comfort in engaging with that the informational asymmetry in terms of us not knowing who the right companies to call, then pulling in a startup, one of the issues with startup, so it always sounds like a good sound bite is just go to Silicon Valley and do this work with Silicon Valley. Well, I’ve done for those companies. The issue is, is that as a company, so here’s where the interests don’t align, right. So let’s get deeper into this because I think this doesn’t get talked about enough. As a customer, the things that I crave, right, I want a mixture of large mixture of stability, cost security things. I don’t want my employee base to be whiplashed by technology every other day with a new thing that comes out and I shoot it at them and they got to retrain themselves and think through how to use it and then deal with security issues. As a CIO or CISO and other people inside our organization. You want stability, you want scale, you want cost, you want security, you want auditability, you want training, you want efficiency, you want productivity. That is completely antithetical to the storyline that the startup is giving, which is about disrupting, disrupting, disrupting, disrupting. Right, it’s the innovators dilemma problem. So on one hand, I want to participate in the excitement of the startup world, while simultaneously creating my operational excellence and efficiency. So somebody told us this to me the other day, you know, it’s, it’s, it’s, it’s probably not cool to quote Elon Musk or use an analogy, I don’t know if it came from him. But the idea of, you know, you’ve got your launches that you’re doing for NASA, that you don’t want to blow up, they need to be 100%, you know, accurate, you want full safety, you don’t want to blow up a multibillion dollar satellite blah. So you want operational efficiency on shooting rockets up. But then you’ve got another test pad where you’re blowing shit up, right? You’re trying out the new engine, you’re trying this thing out. And once you get it to a point of efficiency in that curve, you want to move it from left to right, you want to move it from test to operations. So commingling this idea of like operational stuff with the new stuff, we have to get very smart as a customer, about understanding how to pull innovation in into the right places in the right stuff, in the right time and the right modality, yet at the same time, keeping a long range radar out about that,
Almost as if we gotta be multi dextrous. They know Right?
Right, exactly. Right. And there are people by the way you can hire whose job is to keep the trains running on time, right at scale. And then there are people you bring in who are like out to take down the status. And as a leader, you know, I’m sure director Burns wakes up in the morning is like, how do I keep both of these things going. And a third one for us is it’s not only pulling capability from outside CIA, just like DOD, there’s you know I jokeabout this a lot is there’s no app store for spy software. Right? The stuff we do is so specialized, we can’t get somebody out there. So we do have to get smart about being becoming an integrator, we build a lot of stuff inside, you have to be great and world class of building stuff. So many of the techniques that startups use in terms of how we build software, there are tech right series investing that VCs do, you start small, you work your way up. Typically, in the government, what happens is, unless a project upfront can’t scale to a billion dollars, nobody’s interested in funding it, because it’s just not important enough, because it’s not a billion dollars. The point is, is that we need to get smart about this thing that could turn into a billion dollar TAM, or a product at some point. But we’ll start with $500k or a million and work our way up. So these are kind of like the whole innovation landscape. I hate saying that word because it almost sounds like a meme or like it just sounds trite. Because innovation gets overused. I think everywhere in government to some extent. There’s a lot of tech tourism, a lot of innovation theater, getting down to the basic elements, it’s just like the earlier point, we’ve just got to get smarter about doing it. And we just got to operationalize it and not create this into a big drama and theater.
So Nand you know the pressures founders are under, because you were one. And I think you’re an example that we at BENS push a lot cross-pollination between industry and the US government. So it’s interesting to hear your take on these issues from both perspectives. I’m curious for our listeners that are founders or companies that are trying to break into the intelligence community as well, hopefully doing better what stands out to you when a company comes to see you. What do you want to hear from them?
Oh, yeah. So there’s a couple of like, trite answers I’ll give, which is the classic stuff that we need to do. And I think this at least how we’re, we at CIA, trying to reconfigure ourselves to increase that velocity and match. First and foremost is domain expertise, obviously, an understanding our needs and other things there. We are a super spooky spy agency, we’re not really good about broadcasting our needs, which is terrible. We’re going to try to fix that, at different levels of classification, starting to broadcast out what our needs are an aggregate. And based on kind of what we see and deal with, then we can we can move into higher levels of classification. In other work, we do. There is also an aspect by the way to this whole game, which is a lot of the stuff we do, in some cases doesn’t really need to be classified. Right? If the three of us just decided to go build a spy agency, you know, I think we’re smart enough to know like the 15 to 20 things at the core, we’re gonna need to go do the mission that we do, right? There’s all the stuff we do for ops and things like that, which it’s like super deep or super specific. So that’s separate. The second thing is, is that what I talked about earlier, which is currently the modality of, and this is a deeper issue I think the DOD issue, an us issue, and other issue is this and software issue hardware issue government wide, actually more of a buyside problem, I think just in general generalizing it. What happens is, so let’s talk about the pressures again, you’re a startup CEO, your job as a product company is to build a widget that you can punch off off the production line with zero or low marginal cost, and sell to as many freaking customers as you possibly can. That’s called a business, right? Whether you’re making pizzas, you’re making iPhones, it’s the same thing. Standardization, high scale, low marginal cost, very low cost of sales, no post customer support. That’s like an ideal company. Our problem, though, is we have an incredibly complex infrastructure system, where taking a widget that’s shrink wrapped, and shoving it into this complex environment just doesn’t work. So what ends up happening, it’s the same problem we talked about, even at that panel today was the swivel chair integration problem, where I think widget A, widget B, widget C don’t talk to each other. And I take a human being and have them do swivel chair integration through keyboard and mouse and monitor. So what I would encourage and so so guess what tech startups have sold to the Defense Department and to us, it’s basically vertically encapsulated pieces of hardware, drones, you know, other pieces there, there are very few success stories of software getting sold into the model. Because hardware is something where you can take it, I can give it to a soldier and say go fly this drone. That’s a vertically integrated thing. It doesn’t have dependencies on other things. And then the human being can call out well, I saw this over there, my eyes go put it into a targeting system over a radio, right? That’s crazy, right in today’s day and age. So my message to startups and even the hardware drone providers or the software folks is make your software modular. Turn your software and systems that becomes interface into the other pieces as API calls and platforms. Give me an interface where I can give it to a soldier to pilot the drone, but make an API accessible framework, so I can program the drone. And not I’m trying to get royalties off this paper that general Shanahan and I wrote called Software Defined Warfare. But the basic concept is, is that at some point in time, we’re gonna have to build out very complex battle networks. On the intelligence side, we take in huge data feeds and other things, we need to run them through data processing, and other same problem. It’s all flow of data through applications and systems through multiple levels processing down to a decision maker. Well, if startups came and said, hey, guess what, I can show you the razzle dazzle demo, because everybody loves demos, and it’s like cool user interface. But when I deploy it, you can just program it through through Python.
It’s, it’s it’s almost counterintuitive to the build the Moat model, the way to sell is should not have a moat. That becomes your competitive advantage.
That’s right, your ability to integrate it silo yourself into infrastructure because you know, this idea that like this little widget is going to come. And it’s it’s it’s the one thing that you’re asking for that we’re going to put a human being in front of to go do in today’s day and age. Sounds crazy. It’s this integration problem on the back end. So it’s funny, because this is my second time now on the buy side of software, right? I’ve been on the sell side of software. And every time I would come to a customer, I’d be like, well, here’s my widget, here’s my thing, buy my thing, deploy my thing. I never understood that had empathy with the customer on the other side that said, I got 400 of you coming at me each with a distinct widget. Now, human capital in places like military or even in our place, we viewed it as cheap, right, which is sure I’ll put somebody in front of a keyboard. And they’re not going to quit because they’re on a four year, you know, commitment. So I can have them peeling potatoes, or I can have them doing data entry. And it’s just all the same to me. The imperative in today’s warfare that we’ve all talked about before, which I think this whole conference is all about is as a national security group of technologists, how do we view the changing nature of warfare and battles and intelligence? And how do we build these systems so that it’s not about eliminating humans it’s actually creating speed. It’s creating enablement for our commanders and our senior leaders to do this stuff at scale at speed with the right sort of cost, structure and security that they deserve.
Well Nand I know you have a busy schedule out here today. Thank you so much for sharing these insights many for our listeners to hear from the intelligence community is especially valuable. But again, you’ve spent time at DOD, thank you for weaving in your priorities today, but also advice to folks who are trying to break in and be stronger partners.
Thank you so much. This has been super fun.
You’ve been listening to building the base, a podcast from the business executives for national security. Join hundreds of senior leaders and executives dedicated to the mission of keeping our nation safe. Check out our projects. We’re currently working with important upcoming events and the many ways you can get email@example.com
Building the Base Episode 35: Katherine Boyle
In this week’s episode of Building the Base, hosts Lauren Bedula and Hondo Geurts are joined by Katherine Boyle from Andreessen Horowitz. Katherine, a general partner at Andreessen, leads the…
Episode 22: Stephanie Chung
In Episode 22 of “5 Questions with the CEO,” General Joe Votel talks with trendsetting aviation executive and BENS member, Stephanie Chung. The first African-American president of an aviation brand,…