Jan 17, 2024 · 1h 6m · news
Shyam Sankar: The Broken Incentive Structure of How Governments Buy Defence | E1104 · 20VC with Harry Stebbings
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of 20VC, Harry Stebbings interviews Palantir CTO Shyam Sankar to dissect the broken incentive structures of government defense procurement, contrast it with Palantir's software licensing playbook, and explore Palantir's unique "artist colony" management philosophy alongside the future of artificial intelligence.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Harry holds 19.9% of the talking time here. How this is scored →
Harry as informed peer 3.3 Guest teaching 5.4 Guest disagreement 1.8 Harry pushing back 2.8
speaking balance: gold is Harry, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 53:59 Shyam rejects the VC doctrine that budget constraints enforce creativity
Shyam directly rejects Harry's venture capital axiom that constraints drive creativity, arguing forcefully that financial budgets are inherently anti-creative and must be shielded from talent.
Hardest push from Harry ▶ 53:45 Harry challenges Shyam's artist colony model with layoff realitiesHarry refuses Shyam's idealistic framing of protecting talent, confronting him with the practical realities of corporate budget cuts and forcing 15% workforce layoffs.
Biggest teaching moment ▶ 34:34 Shyam corrects Harry's estimate of U.S. defense spendingWhen Harry asks if U.S. defense spending is around 15% of GDP, Shyam immediately corrects him with precise data showing spending is actually at a historic low of 2-3%.
Harry holds his own ▶ 10:36 Harry challenges Shyam's single-buyer monopsony argumentHarry demonstrates strong market expertise by pointing out that defense prime contractors sell to multiple allied sovereign nations like the UK and France, invalidating Shyam's monopsony premise.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Harry as informed peer | Guest teaching | Guest disagreement | Harry pushing back | Why |
|---|---|---|---|---|---|---|
| Shyam Sankar's Background, Work Ethic, and Philosophy on Luck | 1 | 3 | 1 | 2 | Harry asks background questions about Shyam's father, work ethic, and views on luck. He offers a mild counterpoint at [4:37] suggesting people overvalue 'fake humility' rather than undervaluing hard work. | |
| The Origins of Palantir and the Early Days of Defense Tech | 2 | 4 | 1 | 1 | Shyam recounts joining Zoom (with an X) and Palantir's early days in 2005. Harry demonstrates familiarity with defense tech context, citing Palmer Luckey and Trey Stephens. | |
| The Cost-Plus Trap vs. Palantir's Software Licensing Playbook | 6 | 6 | 2 | 6 | Harry actively interrupts Shyam's premise about government monopsony, arguing that sovereign allies (UK, France, Germany) create a multi-buyer market for high-end defense products. | |
| The 'Land and Expand' Strategy and the Software Delivery Challenge | 3 | 5 | 1 | 1 | Shyam outlines how Palantir's land-and-expand strategy relies on product-driven expansion in air-gapped environments. Harry asks informed follow-ups on hardware versus software procurement mindsets. | |
| The History of America's Defense Industrial Base and 'The Last Supper' | 2 | 7 | 2 | 2 | Shyam provides a historical overview of America's defense industrial base, from World War II industrial mobilization to Bill Perry's 1993 'Last Supper'. Harry asks brief clarifying questions to guide the narrative. | |
| Wartime Agility, the Process Trap, and Kelly Johnson's Legacy | 4 | 6 | 2 | 4 | Shyam references Kelly Johnson and Skunk Works to critique process-heavy procurement. Harry pushes back with industry specifics, noting how director insurance guarantees structurally exclude small innovators. | |
| The Venture Capital Model in Defense and the Fallacy of 'Peanut Butter Spreading' | 4 | 6 | 3 | 3 | Shyam strongly criticizes DoD's 'peanut butter spreading' of capital and labels SBIR small business grants a 'white collar jobs program'. Harry brings up continental protectionism and national purchasing preferences. | |
| Historical Defense Budgets and the Insurance Premium of GDP | 2 | 8 | 2 | 2 | Harry guesses that U.S. defense spending is around 15% of GDP during wartime. Shyam directly educates him that defense spending sits at a historic low of 2-3% of GDP, down from 5-6% during the Cold War. | |
| Battlefield Learning Rates and Attributable Capabilities | 2 | 7 | 2 | 2 | Shyam proposes reforming DoD procurement by introducing internal competition between program managers and testing capabilities in the field. Harry prompts him with a hypothetical leadership scenario. | |
| The Impact of Elections and Budgets on Defense Buying | 3 | 6 | 2 | 1 | Shyam contrasts Washington D.C.'s strict conflict-of-interest avoidance with Silicon Valley's 'no conflict, no interest' philosophy. Harry asks focused questions about continuing resolutions and VC obligations. | |
| The Primacy of Content Over Process in Team Scaling | 4 | 5 | 3 | 6 | When Shyam argues that content supersedes process, Harry challenges the premise by pointing out that process is mandatory when scaling large organizations. | |
| Talent Spotting and Managing an Artist Colony | 3 | 6 | 1 | 2 | Shyam advocates treating tech companies as 'artist colonies' rather than structured factories. Harry steers the conversation into how managers handle talent during non-performing periods. | |
| Managing Underperforming Talent and Budget Constraints | 5 | 5 | 3 | 7 | Harry confronts Shyam with the pragmatic reality of budget cuts and 15% layoffs, questioning if the artist colony model is naive. Shyam rejects the VC trope that financial budget constraints drive creativity. | |
| Transitioning from Factory Hiring to First-Principles Talent Growth | 4 | 4 | 1 | 1 | Shyam shares his 0-for-10 record when importing external corporate managers into Palantir. Harry offers his own contrarian hiring doctrine of prioritizing 'boring people'. | |
| Navigating the Personal Toll of Company Reincarnation | 5 | 5 | 2 | 4 | During the quickfire round, Harry challenges Shyam's prediction of LLM model commoditization by comparing it to historical cloud infrastructure margins. | |
| AI Adoption Hurdles, Final Reflections, and the Future of Palantir | 2 | 4 | 1 | 1 | Shyam warns about AI bubble dynamics when proof-of-concept demos do not yield production value. Harry wraps up with classic reflection questions about leadership and Palantir's five-year outlook. |
Statements from this episode (26)
Defense procurement processes fail at delivering both fairness and speed
“In an effort to have a perception of fair competition, we have created so much process that we either have fair competition nor speed.”
European defense procurement is slower than the US due to fairness mandates
“The European process is even more slow because there's an even greater emphasis on how do I show fairness across all of the EU states.”
US defense spending is at historic lows relative to the Cold War
“We are spending at historic lows relative to the Cold War.”
The commercial sector now conducts the majority of R&D over government
“And so as the world transitioned from a place where all of the R&D was done by the government, fifties, sixties, obviously the preponderance of R&D is done by the commercial world.”
Palantir refused hourly government services revenue to force software license sales
“And so, I think the great opportunity, and really, so the playbook was, look, I'm not gonna let you pay me by the hour. I'm going to refuse to accept any revenue that's services based from you, and instead, I'm gonna build a product, and I'm gonna keep investi…”
Government software expansion is driven by product capabilities, not user licenses
“The first is that the land and expand doesn't work the way it does for commercial SaaS companies. What you're not trying to do is it very much is about expanding the product surface area. So as opposed to the license surface area, or I'm trying to get more use…”
Software R&D costs continuously increase over time, contrary to government expectations
“Software, first of all, software R&D costs only increase. You know, it's like no one, no investor thinks Microsoft's gonna spend less in the future on R&D than they spend in the present moment. That would be crazy. But the government somehow expects that to be…”
US defense prime contractors dropped from 51 to 5 since 1993
“He invited the 51 defense primes. Today we have five. Back then we had 51.”
1993 defense consolidation replaced venture innovation with dividends and stock buybacks
“The much more profound consequence of The Last Supper is it began a deep financialization of defense. It became about dividends and buybacks and leverage ratios. It, you know, it became about M&A. New venture formation was about spin-outs and mergers, not abou…”
Constraining defense procurement to avoid catastrophic failure guarantees mediocre outcomes
“In order to constrain the variance in outcomes, we have constrained ourselves to mediocrity. You know, it's like, we don't want projects to fail catastrophically, but that also means you can't have projects succeed catastrophically, and the amount of process w…”
America has the best defense procurement process when political will aligns
“America actually does have the best process when we're serious. When there's the political will and alignment and the problem is clear and we're ready to go, that's when we throw all the rules and the doctrine out the window and we go and we get it done.”
Roughly $100 billion in venture capital recently entered defense tech
“We've had roughly a hundred billion dollars of venture investment enter defense tech recently, last couple of years.”
The DoD falsely claims it doesn't pick winners despite being a monopsony
“One of the greatest lies they tell themselves is we don't pick winners. They, if you interact with acquisition folks, you'll, you'll get this sort of like, well, we don't pick winners, but no, literally you're a monopsonist. Every contract you award, you're pi…”
Research shows defense companies receiving SBIR grants never achieve scale
“It's research study after study shows that you get these CBER grants and your company never actually achieved scale.”
Many military drones used in Ukraine are single-use, requiring no sustainment
“Many of these drones are single-use drones, so there is no sustainment.”
Battlefield innovation cycles render defense technology obsolete within two years
“What you're gonna fight with in two years is not what you're gonna fight with in a year, and it's not what you'd use today.”
The US previously funded four competing programs for submarine ballistic missiles
“When we were building submarine launch ballistic missiles, we didn't have one program to build it. I think we had four”
The DoD focuses on vendor competition rather than internal program competition
“The government focuses on competition in the industrial base. I think there's not enough competition inside of government.”
Frontline military combatant commanders have no vote in procuring defense equipment
“There is a pretty big disconnect between the people who are responsible for fighting, really the combatant commanders and the people who are responsible for providing the equipment. And really these guys, the fighters, they don't really have a vote.”
The Defense Department cannot launch new programs under congressional continuing resolutions
“In the U S I think the bigger issue is really Congress and whether there's a budget or not, you know, we don't have a budget. We'll enter a continuing resolution in a continuing resolution. That means you can't have any new starts. So all the things the depart…”
Companies should design roles around exceptional talent instead of rigid slots
“Really talented people are highly asymmetric. You're much better off designing a role around who they are, rather than saying, what role can I slot you into?”
Linear corporate career progression is deeply inauthentic for elite talent
“Artists' work don't monotonically get better. This idea that you're gonna climb levels in this sort of strictly increasing way, I think, is deeply inauthentic.”
Managers should shield creative talent from the brutal reality of budgets
“What you really want, I mean, that's the job of the talent manager, is to manage to the budget without exposing the brutal reality of the budget to the talent. You know, how can I get the talent to feel a few less constraints while actually being constrained?”
Palantir survives frequent corporate reinventions by operating like an artist colony
“And I think it's actually only survivable because the whole thing is set up as an artist colony.”
AI value will accrue to infrastructure and apps, squeezing model providers
“I think they'll get a lot of the compute for sure. And that's, and I should say, I think the value is going to accrue in those two directions, infrastructure and app layer, and that the models are going to be squeezed.”
AI model companies must build applications, contrary to standard VC advice
“If I was a model company, I would be pushing into the app layer. I'd be trying to figure out how to own more of the value I'm creating with the model rather than just selling an endpoint. And this is usually the opposite advice you're going to get from traditi…”