#058: Palantir – Global Defense Lead Doug Philippone

Thursday May 05, 2022

Data is the ammunition of the future. What we do with that data, how we understand and interpret it, and the decisions we make with it are what make data the most powerful to ol in the modern day arsenal of both countries and companies. 

Host Fran Racioppi sits down with Doug Philippone, Global Defense Lead for Palantir, to define big data as a decision-making tool. As a former Army Ranger and West Point graduate, Doug started and has led Palantir’s Global Defense division to scale while competing with some of the world’s largest and most institutionalized defense contractors.

Doug shares Palantir’s rise, the hole in the market they’ve exploited, the challenges they’ve faced breaking into the government procurement system, why a technology strategy is not IT Transformation, and how AI and ML are changing the world order.

Listen to the podcast here


About Doug Philippone

TJP 58 | Global Defense LeadDoug Philippone currently serves as the head of Global Defense at Palantir Technologies. Doug both started and successfully grew the defense business to where it now supports over 19 different allied countries. Doug has led every part of the Software Platform business: from providing vision to design and innovation teams that invent new technology to leading the teams fielding those products across a variety of sectors in the world’s harshest environments.

Prior to Palantir, Doug was an Army Ranger who deployed six times and commanded multiple Joint Special Operations Command outstations in support of the Global War on Terror. During his Army career, he was awarded three bronze stars with two valor awards, the joint commendation medal with valor award, as well as many other commendations for his service to the nation. Doug earned a Bachelor of Science degree in Mathematics from the United States Military Academy at West Point, received his Master’s Degree in Terrorism Operations and Finance from the Naval Post Graduate School and studied Machine Learning at MIT. He is also a published author on terrorism finance.


Palantir – Global Defense Lead Doug Philippone

Data is the ammunition for the future. Data determines what we know about ourselves, our adversaries, the environment, our friends, and anything. Collecting data is only the first step. What do we do with that data, how do we understand and interpret it, and the decisions we make with it are what make data the most powerful tool in the modern-day arsenal of both countries and companies.

For this episode, I stepped out of the spring break Florida sun to sit down with Doug Phillippone, Global Defense Lead for Palantir, one of the world’s leading technology companies. Doug and I define big data. We spoke about data as a decision-making tool and why people both embrace and fear data’s ability to provide both information and intelligence.

As a former Army Ranger and West Point graduate, Doug started and has led Palantir’s global defense division to scale as it competed with some of the world’s largest and most institutionalized defense contractors. A David versus Goliath battle that has tested every facet and fiber of his leadership and resolve, Doug shares Palantir’s rise, the hole in the market they’ve exploited, and the challenges they faced breaking into the government procurement system.

Also, why technology strategy is not IT transformation, and how AI and ML are changing the world order. Doug abides by the mantra of a lifetime of service to the nation. He shares some of his new projects as he builds Snowpoint Ventures and continues to invest in public-private partnerships. Data is everywhere. We can fear it or we can embrace it, but we cannot escape it. Doug, welcome to The Jedburgh Podcast.

Thanks for having me.

TJP 58 | Global Defense Lead

Fran & Doug Philippone talk big data, decision-making and a lifetime of service to the nation.

Thank you so much for welcoming us to your home in sunny Florida. It is like spring break here every day. I must credit Dan Rickert. He served with me in ten special forces groups. He works with you now. He has been instrumental in setting this conversation up with you. I do want to give him credit. Thank you, Dan. He didn’t make that joke, though.

Welcome to the free state of Florida.

We spend a lot of time on this show, especially during the first half of 2022, talking about grit and our ability to push through during adverse times. How do we buckle down, find a way to bring calm to chaos, and solve complex challenges that test our character and our values? We first explored grit in terms of personal achievement and professional goals. I’ve expanded this conversation to include a critical discussion on World Order.

It’s important because we are challenged by what I’ve called our frenemies or these near-peer powers in the world like Russia, China, and clear adversaries like Iran and North Korea. Gingrich, who I interviewed, said anyone who seeks to have nuclear weapons now could be considered one of our adversaries.

The conflict in Ukraine has put front and center our premise that leaders and leadership matter now in the world and the World Order more than possibly ever has, as we may be closer to the brink than we’ve ever been. We’ve also spent some time working on this concept of VUCA or the complex world of volatility, uncertainty, complexity and ambiguity. Also, how do we as leaders effectively manage our emotions and our responses and find solutions to problems with no linear resolution.

Problems require not only action but the understanding that when we take an action, there’s probably going to be a 2nd and 3rd order effect that we may or may not be able to understand and be forward-looking to understand how we’re then going to respond to those. Now, we live in a world that is more connected than ever. You know this more than anybody else. At the center of this connectivity sits the word data.

Facts and statistics are collected together for reference or analysis, one definition. Another one, things known or assumed as facts, making the basis of reasoning or calculation. Data, information, intelligence, whatever you want to call it, whether it’s raw or aggregated, allows leaders to make educated and informed decisions to respond to chaos and confusion in the VUCA world.

Palantir sits at the center of big data. Palantir has led data aggregation in the industry in both public and private sectors for the better part of the last decade. I sit here today in my circa 2013 Palantir t-shirt. I have to ask you, though, maybe I’ll do it after you solve this riddle. I got some pictures up. You’re going to need to solve this riddle. In the end, we’re going to circle back. Don’t let me forget.

TJP 58 | Global Defense LeadThat’s what it’s about. That’s what Palantir has done. It has taken data and it has created these frameworks and these platforms to be able to analyze it, triage it, and then allow leaders to make maybe not perfect decisions but more informed decisions. That’s what we seek to do. When I thought about the next conversation of the VUCA world, there was no better person to talk to than you. You’re a former Army Ranger. You have built Palantir as a global defense division. Thank you for joining me.

Thank you. It’s great to have this opportunity. I’m humbled. It feels a little bit surreal as I’ve gotten through my life journey here.

We’re going to talk about it all. There are hundreds of directions that we could go. When we were talking on the phone, we laid out a whole bunch. I’ve done my best in my terms to lay out how I view the importance of data in the world today. You’ve been with Palantir bow for the better part of fourteen years or so, almost the entire lifespan of the company.

You’ve been involved in every aspect of the development of the software. You’ve pushed when others said, “Hold. Don’t push.” You’ve done that internally within the company and you’ve done that externally. Can you describe to me in your words this concept of big data? What does it mean? Why is it so important in the world today?

[bctt tweet=”Why data matters is when it’s turned into information that you can make decisions on.” username=”talentwargroup”]

It’s interesting, when I think about my journey in life and then how that intersected with my time at Palantir, there was a funny moment in the early days. When I joined Palantir, I had to look up data. It was funny because I have a mathematics degree, so in certain circles, I’m thought of as a technical human being. It’s like, “When they say data, what the hell does that mean?” When I transferred that into layman’s terms, whether how raw it is or not, it’s about information in some capacity.

You can get technical in the definition here, but why does data matter? It’s when it’s turned into information that you can make decisions on. In the world that you’re describing that we live in today, as big as it was when I started, it was around this idea that there’s information being collected everywhere. Whether it’s through sensors, computers, or any systems, there’s always something being stored. How do you turn that into something useful? That’s the core if you think about what anybody cares about or most of the tech businesses.

TJP 58 | Global Defense Lead

Doug Philippone: ”Why does data matter? It’s when it’s turned into information that you can make decisions on.”

Why do people fear it? I asked that question because when we think about organizations and companies like Meta, which people are now looking at and saying, “Meta owns all of your data. They own everything about you. We need to fear it.” I don’t get that perspective when I talk to you or I think about Palantir. I think you’re using data for good in a lot of ways.

First, you should understand that we had a completely different approach to all this stuff. When you think of some other companies, your data and personal information is their business. That was a fundamental difference from when Palantir was formed. It was around the ideas. If you think about after 9/11, people don’t remember this, but if you’ve been around for a while, you do, the Patriot Act.

There was this idea that you had to make a binary decision, security and civil liberties because of this trade-off. Our thought was around this idea that you can use technology, so you don’t have to make that choice. How can you protect civil liberties while ensuring safety? The idea of technology, at least at Palantir, was, “One the government was going to screw up, so we need to bring our entrepreneurial spirit to solving these problems so that there’s no trade-off.” We believe that that’s a false choice. You can protect civil liberties and all your information.

Think about normal warrants and so forth. Only the people who are allowed to see this and have a reason to see this can. That was important. It’s the fundamental of like, “We make the architecture that moves the data around and protects it so that you can make informed decisions.” It’s fundamentally different.

If we think about the founding of Palantir by Peter Thiel and Alex Karp in 2004, originally, the algorithm was based on an anti-fraud algorithm from PayPal. After 9/11, there’s an opportunity here to potentially use this for counterterrorism operations. There’s a quote here I want to read and it’s in 2004, “When we looked at the available technology, we saw too rigid products to handle novel problems. Custom systems that took too long to deploy and require too many services to maintain and improve.”

To many extents, this is a battle that you’re still fighting today. One of the core foundational elements of Palantir is they say, “We believe in augmenting human intelligence, not replacing it.” Entrepreneurism and the entrepreneurial spirit and drive are important parts of our conversations on a podcast.

Palantir launched, took on, and entrenched defense industry behemoths Raytheon, Lockheed Martin, Northrop Grumman, Boeing. Since that point in time, you have been at what I call in the middle of the one-meter nice light. Can you talk about the hole in the market that you saw when you looked at the big defense contractors, how they were looking at the data problem, and the customization of the products they were creating? Also, how Palantir could come into the market and pioneer a new way to think.

In full disclosure, in 2004 and 2005, I was on the Syrian border.

You were still in the Army.

I wasn’t JSOC at that point. That was the start of my transition over there. It’s even more foundational now. This is more of a problem now than it even was then and it was a big problem then. You fundamentally didn’t move data around between systems and have any ability to understand the pedigree of the information or the security of that information before it would move around.

You have these problems where you have one system that works. You have a bunch of systems that either work or don’t work, but there was never a holistic enterprise that could move the information throughout an agency, the safeguards, or certainly not between agencies. This is even a worse problem now.

It’s probably one of many reasons why Palantir has been so successful in both commercial markets and broader government around the world. Most people don’t understand this and they shouldn’t. Between making an app, iPhone apps or something like that, or in the sense of what’s called enterprise software, the difference in complexity is massive. You could go to school and you don’t even need a college degree. You just have to be a generally smart human being, and I could train you for a month and you could build some apps. There’s a relatively low bar to build some apps.

There are no-code apps that you can build them.

They’re like macros. There’s enough infrastructure around in the world where you can do that. What does it mean to have enterprise software? That’s one of the hardest things out there. By most metrics, you can read this 93%-plus failure rate on those types of projects. When people have figured out how to do this, they’re the biggest companies in the world. Who tries to make the most enterprise software? The largest companies in the world and the government.

Most of those systems are mostly failures. The problem is a lack of accountability in one way, but mostly, it’s just changing the name and keeping going on. Tackling that problem of, in the early days, can we get a bunch of women? You think of this as you’re trying to recruit the smartest engineers in the world.

Can you get a working system that works first to solve this problem and then go to market with that? The quote that you made versus when I started was a four-year difference. You’re not selling to anybody. You’re just building something that works before you even go to market versus a PowerPoint promise that then you’re going to say you’re going to build it, but then the risk of failure is massive.

I remember my first experience with Palantir was in Africa in 2013. They had been around for a while. I remember loosely hearing about it. We’re in East Africa and we know that we don’t know a lot about the continent. You have your normal intelligence collection efforts through the embassies and the different organizations that are out there.

You have military units and special forces teams throughout the region, but there was nothing where we could look at and be like, “I have a sense of what’s happening on this complex continent where everything’s trying to kill you, including the land itself at times.” It was our joke. We put this challenge back up to our headquarters and said, “What can we do?”

Lo and behold, a couple of weeks later, these two guys show up with a box of Palantir t-shirts and a couple of Panasonic Toughbooks and they’re like, “We need ten TVs in a conference room.” “First, I don’t have ten TVs. Second, we’re living in a container here. I don’t have a conference room. You can have maybe that little closet over there and three TVs. What are you doing?”

TJP 58 | Global Defense Lead

The Jedburgh Podcast Episode 58: Palantir Global Defense Lead Doug Philippone

They ended up creating what they call this monitoring center. We were like, “I don’t know what that does. Eventually, maybe I’ll come to talk to you, but I got other things to do.” There was a coup in March of 2013 in the Central African Republic. We had teams who were in and about the area. We had teams working with the Lord’s Resistance Army effort.

All of a sudden, it’s like, “Those guys in there might know what’s going on,” because they’re aggregating all other stuff. Now here we are, the entire staff sitting in this room with these two guys, giving us a lot more understanding of what was going on in the environment. For me, that was a huge eye-opening experience as to what some level of data aggregation can do. The company has advanced tremendously, leaps and bounds beyond that, but from an early experience, that’s my Palantir story.

First, it’s great to hear that it helps you in the real world because that’s what we’re trying to do. It’s progressed a lot. Some people don’t understand the idea that software is never done. Every day, we have 2,500 or so engineers and they’re waking up and building new boundaries every day. That’s informed by all of the things that we’ve seen over the years and a variety of problems about either the industries or the problems you see, the different data you’re seeing around the world.

This is what appealed to me most when I left the military and joined Palantir. It was this idea of lifetime service. How can you continue to serve in some way in a different capacity after your body breaks? I can’t jump out of airplanes anymore, but I could lean on my math degree and go do something useful for the world.

As a commander, the idea for me was trying to figure out what was going on and making good decisions. From a technical perspective, the challenge is that the majority of the world works because people aggregate information and then put it in flat formats. Flat formats would be like putting that in a Microsoft Word or PowerPoint thing and it’s a static thing that doesn’t send a living, breathing thing. I’m presenting it to you and it has a timestamp on it, “This information is as of,” a time date group, then they’re like, “This is what we think.”

[bctt tweet=”Software is never done.” username=”talentwargroup”]

That’s what they teach you to put on there, and how long is it a value.

The interesting from a technical perspective is this idea of how do you change this paradigm? It’s a little bit of a joke because, generally, institutions are a failure at this, where it’s the idea of knowledge management. What does that mean? Typically, from my experience working with the government or the military, it was this idea of, I have these file folders that were written on PowerPoint slides or there are maps or emails or something that you had collected, and then you put it in a hard drive. As you leave, you hand it off to somebody else, and then they may or may not read it, or whatever.

Most likely not.

It’s hard to find. You don’t have this living, breathing thing. What we found is a lot of the lack of progress was this idea that you’re not learning from the past. You’re not improving every day. You’re not improving based on the knowledge of other people. You’re forgetting and then you’re having to learn it again. Afghanistan is over, but the idea is during that twenty-plus year period or whatever it was, we probably fought 40 wars every six months. You find it over. It’s like, “If I only would have known.”

That was the worst. We would switch with the Fifth Group in Iraq. We would come in and immediately think everything they did was wrong. We would do it our way. They would come in, spend two weeks telling us how everything we did was wrong, and do it their way. We come back six months later and have the same conversation. It was exactly what happened. You continue to fight the same war over and over again.

TJP 58 | Global Defense Lead

The Jedburgh Podcast Episode 58: Palantir Global Defense Lead Doug Philippone

The problem is understanding what people need. If you think about even being a politician in many of these areas, you know this better than I do as a former special forces operator, is the idea of the local populace or the villages. The local elder, where do they live? What do they care about? What’s the school? What’s the water supply? There’s a pump. What did I give them money for? It’s so that they could farm or something.

The new unit comes and you don’t have a record of like, “There’s a dam project. The elder cares about that.” You lose all this stuff. There’s huge fraud and waste. We had a special IG for Afghanistan reconstruction and Iraqi reconstruction where we’re seeing the billions of dollars that we wasted or ended up on a plane that went to Dubai.

All this avoidable stuff. I’m giving you a broader background of the problem and I haven’t hit the solution. The idea here is, from my perspective as a commander, my final tours were on the Syrian border and then in Ramadi when it was ugly still. You started to see when I saw Palantir very early, which is funny in retrospect how much it’s changed. It’s not even the same thing.

The idea of, “How could I have stored this information in a meaningful way so that I could search and discover it and find it with a pedigree?” One of the problems with these PowerPoint slides, at least in the military and intelligence context, is that it loses the pedigree of where did this information come from? Why do you know it?

This is a very critical thing for civil liberties as well. From a technical perspective, if you don’t have the pipeline of, where’s the raw source of this information? It could be an interview you did that’s handwritten and becomes typed into a thing. It could be video, some ISR camera, electronic intercepts, or information from the internet. It could be a plethora of data sources. There are thousands of these things we’ll call the data source. That can mean any of those things I said. Understand where those things live.

There’s this term I took from a book I read that was funny. It’s a fake word called the complected system, the idea of when you have these independent column apps. I went back to my earlier comment of how easy these things are to make and this is where the government struggles. Sometimes you have these independent capabilities that both work and are useful, but they interact with data or information separately, independently.

You have all these data sources, then you have all these tools or the apps that do something useful for you. Those independently sit on all the data sources. What you end up doing is building a platform one-off forever. You’re like, “This tool number A interacts with a data source A, B, C, D.” You then start to think about the complexity of that.

How do you protect the lineage of that? Who’s allowed to see it? You have access control. You must do the next tool. How are these two tools interacting with each other? As the operator, how do you log into a website? You have to go to each little thing. That’s how the government is doing it. That was the landscape of how they were like, “The world didn’t know it at the time, but Palantir was really onto one of the most valuable problems in the world.”

If you solve it, how do you build a platform that lies on top of all the data sources? N plus one. It could go to infinity. It could be the old IBM mainframe from the ‘50s or it could be the most cutting-edge AI sensor now, anything in between. It could be different data, file formats, and so forth. Normalize the data. It’s called ontology.

When you model the world in the data formats from before, it could be from 50 years ago, so you call it a car, but there’s no sense of a car stereo. Now, you have all these different terms or have backup cameras. We didn’t have that. That’s a funny example of how the world changes. The world changes all the time. You don’t call things the same as you used to call them. If you have some rigidly locked schema from fifty years ago, that becomes a technical challenge of how do you normalize you’re calling these things suddenly the same thing or one-off? Over time, this may become what’s called complected.

If you want to think of a rope and all the twine of the rope interlocking, when you have one, it becomes simple. When you make one change, it affects every other twine in that rope. That is the difference between apps and enterprise software where you have all these interdependencies like, “I want to do some quick tweak to this stuff.”

You must start thinking about when you make that change to the front end, what you look at, and how that affects everything in the entire enterprise? It gets complicated, and then that’s where these big systems fail. Long story short that was the hardest technical challenge. If you want to make this information useful for your decision, it takes a while to get there.

Let’s talk about you. West Point graduate in Math, a class president, Master’s in Terrorism Operations and Finance from the Naval Postgraduate School, studied machine learning at MIT, six deployments to Iraq, Afghanistan, and Pakistan between 2003 to 2007. You served in Ranger Regiment the majority of that time, three bronze stars, two with Valor, a Joint Commendation Medal with Valor, generally known as an innovator and a disrupter who brings critical thinking to complex challenges. Why go into the army in the first place?

I enlisted in the Army. I went to a fancy high school in Arizona. I assumed it was like you’re on the treadmill. Everyone’s going to college and it’s what you’re doing.

No choice.

I got on the conveyor belts to the lemming farm. I did about a year and I was a complete failure in college for the first time. I had always been into the military and I thought about serving. My family goes way back. No one was a professional soldier but going back to my great grandfather, who served in World War I, my grandfather was in World War II, and my dad served during Vietnam. He didn’t go to Vietnam. Everybody had served.

It was a little bit of how you rebel against your parents. My parents were hippies, so I was a young Republican. You do the opposite. I was like, “I’m going to join the military,” and my mom cried. I wanted to do something super challenging. I got the idea. I had a Ranger contract, so I enlisted and went into First Ranger Battalion after I passed all the tests and stuff. Anyway, it was about doing something hard. When I was in Ranger school, I was a private first class at the time. All these friends that I had made in Ranger school told me, “You should try to go to West Point.”

When I graduated from Ranger school, I went back to this Hunter Army Airfield base in Savannah. I went to the education center and took the SATs and applied. I was going to continue my army career. At the very last second in May, some smart kid probably went to Harvard or something and they had one spot and gave it to the Ranger kid. I got to go to West Point. I found out in early May or something, so it was probably like, “Give him a chance.”

You were right into it. After West Point, you went back and continued to serve in JSOC.

I was an officer back in the infantry. I love the Rangers, so I wanted to go back to the Rangers. When I had done my time, I applied back to the Rangers and got in. Every time I was eligible as an officer, I went back until I got hurt and was medically retired. My last job was the S3 at the special troops’ battalion in Raiders.

Touch about some of the leadership that comes out of that. You’ve transitioned out and got out in 2008. You’ve now gone into building this department and this massive division within Palantir. When you think about what you take from your time in the Ranger regiment, what are some of the core principles that apply?

[bctt tweet=”If you do something useful or you’re producing outcomes, people will leave you alone.” username=”talentwargroup”]

It’s interesting. Anybody who has been friends with me during this time has always known me as a maverick of sorts, or a rebel is a more pejorative term, troublemaker. Palantir is thought of as the most rebellious company and then I was a rebel within Palantir, which is funny.

It’s what gets things done. No one shakes it up.

I had a very different military career, to be clear. At West Point, even though I was class president, I was always getting in trouble, but I had a different career in the Rangers. Whether I was at the right place at the right time throughout my entire life, which seems like it, it was also about taking advantage of opportunities.

TJP 58 | Global Defense Lead

Doug Philippone: ”You’re seldom trained for the thing that you are having to do.”

If I think about the most important things that I learned through my military experiences, it’s mostly formed by the war more than anything. You are seldom trained for the thing that you were having to do. It’s not like we weren’t training. We’re training all the time. You were thrown into these things where you had to figure it out.

For a lot of reasons, JSOC was spread super thin, so you are oftentimes elevated way above your paygrade and doing super complicated things with no adult supervision. All of a sudden, I was the adult and I’m like, “I’ve got to figure this out.” It’s an enormous responsibility. Many times, I was out on the edge of the moon by myself.

This is where I became friends with Stan McChrystal. As a commander, you’re out there and you have to figure it out. You have some high-level intent from a three-star general who’s four hours away. You’re talking to him on a satellite conference call or something like that. You have to understand what’s the highest-level intent? On the ground, I have to figure this out with my folks.

This is a bit of a longer story, but the important part of that is I felt like my experience in war in very nebulous environments where your life depends on it and you have to win. It defined my whole life. Coming into a startup community, that part felt very natural to me because there’s chaos all around and you have to figure it out.

Dan and I talked about this a lot. When I started Palantir, the first three days, we had this getting started guide thing, even following this list of installing servers, doing all this stuff. I was like, “I’m going to get fired on day three.” I’m googling data. I was like, “I need to figure out my path here because if I let them just tell me what to do, they’re going to probably tell me something stupid. They’re not going to know how to take advantage of my skills. I need to figure this out because I’m rebuilding my life here. I know that I can add value and I can be the best person, but it’s not what they’re going to tell me to do.”

About that time, I remember telling Sean, our COO, “I’m going to stop doing this and I’m going to do something else.” The great thing about Palantir is being the perfect place for true entrepreneurs, but specifically for me, I stopped following the plan on day three and never looked back. It was this idea where if you do something useful or you’re producing outcomes, people will leave you alone. It’s the idea of, they’ll pile resources into you where you can hire people that you could do things or support you. You just need to keep kicking ass.

The question goes from what are you doing to what do you need.

There was a certain time during my career where internally, there was a lot of friction like, “What the hell is Doug doing?” I think they could still say that now. I’ve always had amazing support from the leadership. We’re investing. I always had carte blanche, but it was like, “As long as I don’t screw up that trust and I deliver, they’ll support.” That part of it has been amazing.

From a startup perspective, figuring out how to win in war, the closest that a civilian will come to war is living in the startup world where you have to figure it out. You have a limited amount of money and you have to build a product, sell a product, and then your family’s lives depend on it. You have to feed families. That’s important.

TJP 58 | Global Defense Lead

Doug Philippone: ”The closest that a civilian will come to war is living in the startup world where you have to figure it out.”

You brought up General McChrystal. I want to ask you a question about him because I believe that one of the biggest things you take away from leadership in the military is the classification of risk and risk-averse opportunity and the analysis that you must make as a decision-maker. Data, information intelligence, you’ve mentioned them all. They’re tools that leaders use to better understand the environment, classify risk levels, and then decide based on that.

That applies not only in the military but to the startup world. That applies even to Fortune 500 companies that have been around for 150 years. It’s all about the risk categorization and then the inputs that they’re using to make those decisions. We had General McChrystal in episode 34. We talked about his Risk immune system and how leaders in organizations detect, assess, respond, and learn from Risk.

I told you I wasn’t going to quiz you on those. I’m interested in your thoughts on how you create an environment and keep that environment going over time as you have within this organization. You encourage subordinate leaders to take risks and act decisively even when they don’t have all the information that they need to make the best decisions. They must make the right decisions based on the time and place they’re at.

I ask you this question because you have brought up in several conversations that you’ve had external to this one, the fact that leaders have to act, people have to take action. You can’t sit back and wait for everything to be perfect. If you do that, then you’re missing opportunities. When you think about creating an environment in an organization where people are encouraged to take action, how do you maintain that over time?

I’m aggressive. There is a technical answer. There’s a Palantir answer for this, then like, “How do Doug and Palantir the same thing? How are these morphs together?” From the technical Palantir side, the idea of understanding what’s going on but even knowing what’s going on at all, what decision are you making? How does it relate to the real world? Do you even have accurate information? How do you assess the source of that information so you can understand because otherwise, you’re making poor decisions?

It’s not because you aren’t trying. You just don’t have all the information. That’s the challenge of modern warfare. You’re trying to make the best decision as quickly as you can. Sometimes you can wait three weeks and make a better decision, but you don’t have that time. You’re like, “Based on the information I have right now, what’s the best decision I can make?”

TJP 58 | Global Defense Lead

Doug Philippone: ”That’s the challenge of modern warfare. You’re trying to make the best decision as quick as you can.”

How do you even calculate the risk? What’s the downside of this?” Many of the things that I work on in this are a combination of my Math background and Palantir versus the real world. There’s a thing called constrained optimization, where you’re trying to optimize one thing based on a lot of different variables that all have different tradeoffs.

It’s called the Pareto optimal solution. How do you get the best solution given all of these constraints? You figure that out quickly. Sometimes you don’t understand the cost of the decisions. Sometimes you’ll have this false dilemma of where you presented these decisions and then it’s impossible to understand the risk because you don’t know the 2nd or 3rd order effect of those decisions.

From what we try to do with Palantir, we have solutions for this. It’s like, “How do you know the third? When you make this decision, what’s the fourth order effect that’s going to be for the person in East Africa? What’s that decision? Resource allocation, you don’t even know what this means. This is one side of it.

We’re working hard on a technical solution of how do you let commanders understand as many of the inputs you have and the cost of those decisions, and you can calculate the risk of that and put that to the forefront as much as possible? The second part, from a more personal perspective, is the idea of doing something and being bold. You can always make fun of lawyers all the time, but their default mode is to do nothing.

Wait to see what happens. Maybe they’ll go away.

Go back to sleep.

If you want to know, bring in the lawyer and the HR guy. I work in HR, so I can say that. Now all lawyers will send me hate mail.

Sometimes I have this conversation with Sean, but it’s a joke like, “If you want me to go back to sleep, that’s easy.” I’m talking about doing something and doing something always has risks. If I think of all the ways, especially when you’re trying to be an entrepreneur and you’re trying to change the status quo that’s broken, you have to be bold and break a little glass. You don’t have to be a jerk about it.

[bctt tweet=”Sometimes you don’t understand the cost of your decisions. It’s impossible to understand the risk because you don’t know the second and third order effect of those decisions.” username=”talentwargroup”]

By definition, you’re changing the status quo and you’re messing with other people’s rice balls. They get protective. What I had to find out the hard way when I think about the beginning of Palantir, we’re all naive and we’re like, “We’re going to make this great product that works.” That’s not a given. It works. People like it and it’s cheaper. We’re like, “This is going to be so easy.”

Next thing you know, it was the frozen middle, the bureaucrats, whatever you want to call them, moving heaven and earth. The one time that a bureaucracy is agile is while defending itself. They were like, “We’re going to stop Palantir.” We were upsetting the whole status quo. I was like, “This is harder than fighting Al-Qaeda.”

In many ways, the idea of we wouldn’t have won our battle if I hadn’t been at war and been able to stare at the dragon’s mouth. There are times when I want to quit. I flipped in New York one time. I was riding in a vehicle with Dr. Carper and I was like, “Sir, do you still want me to do this? Because this is super hard.” I’m not convinced that any amount of hard work or talent will equal victory.

The defining moment of leadership for me, he gave me what I needed. He said, “If any soldiers need this or want this, you should keep fighting. It’s too important.” I was like, “I could work with that.” I got out of the SUV and I was like, “Back in the final deck now. I have to do this.” One week later, we had some minor victories on the frontier. I was like, “Great. The campaign is still going. I have enough tactical success to be like, “I can win this,” because, at times, it can be overwhelming. I don’t know if I can be quiet. It’s important to do something. You have to do something.

Let’s talk about that battleground. That battleground comes down to what you’ve called this battle between technology strategy versus IT transformation. How does Palantir look at it? How do you look at it as it is on the side of technology strategy versus the government procurement system or institutional legacy defense contractors or technology builders who say that it’s an IT transformation? Can you define those two and talk about that divide?

Large corporations end up looking like the government even though they all pretend like they aren’t anything like the government. Even within Palantir, we have a commercial wing and a government wing. Early on, the commercial folks were all like, “It’s going to be so much easier. They weren’t so nimble.” They’ve called me a few times and asked for my advice.

You see the same paradigm, whether it’s the commercial side or the government side, of this thing of we’re going to pontificate and we’re going to write a strategy paper of how we’re going to transform and we’re going to talk about it. We’re going to do a data cataloging thing or we’re going to have some grand transformation thing.

You have to go do something and start somewhere. This is an idea of solving a problem. There’s this idea where you have this paralysis by study. The government’s the worst at this. It’s like, “Let me do a three-year study. If you look at JADC2, the Joint All-Domain Command and Control, it’s this large cyber thing. It’s a framework to build an implementation around.

In the joint community, the different services can talk to each other, exchange information, command and control forces, and win wars. The idea of that is, “Let’s study this and understand the system.” There’s writing about it. Congress is begging them to do it. That is an example of these large projects where nothing happens. In the government frame, there’s a problem with the way that the requirements and acquisition process.

TJP 58 | Global Defense Lead

Doug Philippone: ”If you do something useful or you’re producing outcomes, people will leave you alone.”

A service would say, “I have a requirement to do something,” and then you’re like, “We’re going to have a requirements generator. We’re going to list down all the requirements you have.” They’re going to have an acquisition community to say, “I’m going to validate these requirements and you’re going to build an acquisition approach.” You’re going to solicit whatever you’re going to solicit from vendors like, “Can we do this?” You feel it.

Five years later, it doesn’t work as advertised in the firm, as opposed to how you can rapidly iterate and give them something, then modify that something? How do you get into that dynamic of getting going with something? This is a fundamental problem with the acquisition. I don’t know if we’re going to solve it in our lifetime. We’ve seen some changes in the system around the idea of prototypes.

They have these things called Other Transaction Authority, which is this old government thing. It’s from the ‘60s from NASA. It was a legal framework where acquisition people could feel prototypes quicker. It’s been successful. From our perspective at Palantir, this is the whole venture community. This is where the government needs to lean in on this because otherwise, you’re not going to get entrepreneurs. Entrepreneurs can’t break in.

The idea is that if you have a better mousetrap, you show up with a prototype. If it works, then you compete with other things that work. That’s the most important thing. When you think about Palantir’s battle to break in, it was this idea that initially, when I go back to 2008, we weren’t competing with real things. We were competing with PowerPoint.

In many ways, we’re still around the world. We’re competing with a CIO or CTO, somebody who’s talking about a thing that they don’t have and they’re going to build it. It’s going to be better and the organization will own it or the government will own it. It’s going to be cheaper. It never shows up and it always costs ten times more.

This idea from the entrepreneur spirit where people can show up with a real thing where they’ve taken investors’ capital and built something cool that can help solve the problem. You show up, it’s a prototype, and it works. When Palantir was about to start in the military, we didn’t know what we didn’t know, but we’re like, “Let’s show this to somebody,” and they’ll give you feedback on if it’s useful. They’d be like, “Do you want to buy this?” They usually say yes, then we’d have to figure out how to buy for the next six months. We didn’t know.

Solve that problem when you get there. This acquisition process is something that you’ve brought up in a couple of different contexts. We’ve had a conversation about it, too, in episode 35. We talked to retired general Craig Whelden. He’s the president of a company called Velontra Hypersonic. He referenced Hypersonic in a variety of conversations because this battlefield of Hypersonic is coming to the forefront.

This is a technology that hasn’t been iterated on since the 1960s with the SR-71 Blackbird, which was then retired in 1999. Yet here we are saying that we have Hypersonic technology that’s decades and decades old. China is advancing this technology at a rapid pace. You talked about how do you get ahead of a Hypersonic weapon? You don’t allow it to launch.

If you look at the government procurement process where they’re building these customizable tools, if we think about the Blue Force tracker that we had in all our vehicles and primarily, we unplugged it because it drained the battery. I set it again and it never worked. Every time you needed it to operate, you had to call a contractor who had to come and you’re looking at this person like, “You’re making $1,000 a day sitting in our back, fixing this piece of equipment.”

I used to carry a Garmin GPS that worked better than the Blue Force tracker in Delhi, where I was. You’ve got a solution that is sitting here at 80% and it’s scalable. When you talk about agility, do you think that the government and the procurement process can get to a point where they can see that light? Can you give a product that everybody looks at, but on the other side of that, they have to be willing to accept it and break down these institutional walls of the Goliaths that you talked about? How do you get there? I know that that’s a long question.

I forgot when we had the lawsuit with the government. At some point during Palantir’s career, we were shut out completely where we couldn’t even compete. There are lots of what they call protests, but 80% of every acquisition type thing is protested in some capacity. It’s normal. Normally, we didn’t think that the competition was fair or we lost. We aren’t complaining about it. We just want to overturn that thing.

In our case, when we were fighting against a break-in, there was no on-road path that we could figure out. We knew that there was this big friction between soldiers and operators who wanted our platform for reasons we didn’t understand. They were forbidden from buying. The existing system wasn’t working and there wasn’t an off-ramp, and there was no way to get there from there.”

If your current stuff doesn’t work, you don’t want to buy it from us. It was crazy.

Finally, they said, “We’re going to recompete this.” They wrote the competition in a strange language where we couldn’t compete because it said that they were going to build. It’s called a pre-award protest, which is unusual because they didn’t award it to anybody. We’re saying, “This is written to exclude us. Why would you do that?”

Long story short, a path of my life journey was for six months or more, I had to become a corporate lawyer. I went to the Court of Federal Claims, the Court of Appeals, and so forth. We won both of those things. I had to dive into the law. There is a law called USC 2377, the commercial item preference. Depending on how old you are, during the ‘90s, it used to be the preference of the government in the Department of Defense. The preference used to be bespoke to everything, which is interesting. Like military spec, the idea, that’s a term. The default mode was to invent everything on their own.

In the early ‘90s, there was this big scandal around the $400 hammers, $200 toilet seats, and whatever. You can google this if you haven’t heard of this. At that point, they changed the law. They said, “There is a commercial item preference for the government. You must buy commercials first.” The idea is, “If you need something, go to Home Depot and go buy the hammer there instead of having Raytheon build the $400 hammer.

I was a mechanized infantry platoon leader in 4th ID. That command inspection of what they call the BII, all the hammers, screwdrivers, and everything, was the absolute most stressful thing you did because you couldn’t go at that point down to the Home Depot if you couldn’t go at that point down to the Home Depot you lost it. You were sitting here saying, “If I lose this thing, I owe you $400 for a hammer and the thing cost me $15 at Home Depot.”

[bctt tweet=”Doing something always has risk.” username=”talentwargroup”]

It was crazy. Where this applies is very important to software. When you ask a main prime integrator to build up a car or tank or something like that, like a Humvee or NI tank or something, they’re pretty good about that. They’re still expensive, but it works. It does the stuff. You can tell, “It has three wheels. It doesn’t work.”

With software, it’s harder to tell if it works. There’s this mismatch between can a casual observer detect whether this works or not? In the software procurement space, in particular, even though this law had been on the books since the ‘90s, the preference was always services contracts of, “I’m going to have an army of contractors that are going to sit here and they’re going to develop this stuff on scratch.” It almost never works. Ninety-three percent of the time, it literally fails and then they have to change the name.

It became this huge problem. Our case was the first time that the law had been enforced in almost twenty-plus years of it being around. It set a precedent, which is great. It’s a huge victory. It’s a three-part task. When you have a problem, the first thing you should ask is, “Does anybody make this?” You put out what’s called market research or whatever, but ask a simple question, “Can anybody do this?”

You give a certain amount of time and people raise their hand and say, “Yeah, we can do this.” They’re like, “Show it to me,” to see that it works. If it was a commercial thing, a commercial item, which means as a human being bought it in the world, there’s a presumption that it works. In that capacity, if you can show up, going back to your risk thing, you’re eliminating for the government and the agency the risk of failure by doing that.

TJP 58 | Global Defense Lead

Doug Philippone: ”Err on the side of just doing some and be bold.”

The first part of the thing is that if there’s a commercial item meets your requirements, buy it and move on with your life. The second part of it is if it doesn’t meet all your requirements, could it meet some of your requirements? Can you just buy that thing and then modify it? That’s the second part. The third most stringent thing is, and this is the real presumption. If it doesn’t, can you just modify your requirements so that you would buy the most commercial stuff first? If you have to do a bespoke thing, it’s a minimal part of this because they always know that it’s a high risk of failure and it’s expensive.

That’s the ground. That’s 4 or 5 years in the rearview mirror at this point. There is still a frozen middle who didn’t get the memo and they don’t know that that’s still around. You still see little pockets of bureaucrats or acquisition people who were like, “No, I’m going to build this. It is going to be amazing.” It’s a fool’s errand, but it is what it is. You fight skirmishes around them.

As a whole, we’re seeing great movement in the Space Force, Air Force Army. They got the memo and they understand. This is important for our nation. In software, in particular, we’re the best in the world. The second is ten levels behind us. It gets scary if we use that advantage and everyone is taking their talent. Our best engineers are going to go sell ads or make video games. That’s a problem.

What we want to do is how do we get those folks to work on government problems? To get that ecosystem going with Palantir, how do we hire? How do we get those people to want to work on these things? The good news is that a decent amount of people want to do this. A lot of patriotic people understand how important this is.

Everyone wants a good functioning government. They mostly want security. There are different levels of people understanding what it means to have security and take it for granted or not. I guarantee you 100% of people want to function in government, whether they understand international security or what that means. We have to foster that environment. To do that, you have to get those right people working on it. My plea to the government is you have to let people win. You have to have an environment where you want to do this, so they don’t just go build video games.

The talent management piece is a critical part of building the company, but it’s building the technology that has these strategic effects. In SOF, we talk about the number one SOF truth. People are more important than hardware. If we don’t have the right people with the right skills to apply them to the right problem at the right time, how do you have strategic implications?

That ties very well into this conversation about where we are with respect to the World Order today. I told you I was going to bring in this concept that I spoke with Newt Gingrich about, which was the elements of national power, diplomatic, informational, military and economic. Doctrinal term, if you remember from back in the day. I didn’t dig that one out in preparation for that conversation.

I bring it up because you said that data was the new ammunition for the military. When you think about the elements of national power and think about how a country asserts its influence globally, it comes down to these four core principles. You still can’t do any of that without data. When you say that data is the new ammunition for the military, what does that mean?

If you think about it, you’re trying to figure out what’s going on. This is all about decision-making. This applies to every commercial organization in the world, but specifically, this applies to our government, our military, and informing our civilian leaders to make the best decision possible. We have the best airports in the world by order of magnitude. We have a competitive space race.

If you think about whether it’s UAVs flying around and you think of drones or you think of satellites, depending on what part of the world you’re talking about, you have limited ability. It’s a contested space where you can’t fly a drone over Moscow. Back in the day, if we tried to fly U-2s over there, they’d get shot down. There are certain costs. If you want to figure out what’s going on, you need to have some sensor that’s going to figure it out.

Going to the hypersonic question, you have to get left of the bang. You have to make important decisions before this happens or else you’ve already lost. How do you figure out what’s going on? When you think about data as the ammunition, if you’re going to have a plane in the air, you’re going to have some electronic sensor that’s intercepting signals. You have to think about all the ways that we collect information to make good decisions. This is our military-industrial complex. The DOD and the intelligence community are helping inform our leaders.

TJP 58 | Palantir Global Defense LeadIf you think of satellites that are flying around, they’re collecting information. Whether it’s electro-optical, synthetic aperture radar, or looking through clouds, how do we know what’s going on on the ground? Is this fake? Is this real? How do we know it? What’s the pedigree of that information? How do you move around? This becomes a problem. You’re collecting so much information. You talk about this big data term. There’s so much information. How do you process it in a way that you can make a meaningful decision in the time period where it still matters?

You think of it as, “This is going to go to the basement of Langley or the Pentagon. In six months, I’m going to get some PowerPoint slides.” That’s always the problem. Think of your time as a commander. You have all these national-level assets. Are they available to you? No. Sometimes there are such a plethora of systems out there where you’re on a classified network and you have to go to 50 different websites to even find out where these things are and then you have to download it.

You’re then going to make your own PowerPoint slide and then send it. You’re a commander and you send in a request for information and it takes you three weeks to get something back. That’s not useful anymore because the events have already happened. That’s an example of not having ammunition. You don’t have the information. You didn’t know. How can you get that information? This is about getting it to the commanders.

You talk about this term called the strategic corporal. You heard this term where it’s a low-ranking person who has strategic importance because they’re at the pointy end of the spear where it matters. This is a good thing and a bad thing. These lower-level players become political and strategic in nature, but they don’t have the same tools that the president would have. I’m not saying there are reasons that you don’t want them to have this as well.

The long story short is, how can you get as much information that somebody is allowed to see to the lowest level possible? What Palantir is working on every day is how do we move all this information around so that it gets to people so they can make the decision? That is an example of that. The ammunition is the lifeblood. If you can’t make that decision, you could have significant implications.

You’re never taking the human out of the loop. You said that you’re not a fan of taking humans out of the loop. Ai is a big component of what you guys are working with. You’ve said that AI must be functional and it must be useful. I think about AI in a lot of ways. Leadership. There are a lot of people who talk about it and there are few who know what it means.

The job of AI is to collate information and then make these recommendations for decision-makers, as you’ve said. When you look forward and look at how we can now apply all of these tools to this strategic corporal, how do you see the future of AI? What’s next for that battlefield corporal and, first, the president?

You’re right in the sense that most people don’t understand what AI is. The term AI can be so broad and misused that it’s almost like a fake term.

People say, “AI and ML.” It’s like, “I know what I’m talking about. I said AI and ML.” “The algorithm does it.” “The algorithm does what?” “It.”

Machine learning, in terms of art, is a real thing. You can find it functioning through. When I make the term AI and ML, it is important in terms of most AI doesn’t work at all. It’s two-step forward. You have to have an algorithmic approach that does something and works and the thing that works has to be useful. This is where the broader mission in Palantir feeds into the AI universe and makes it work.

Many people don’t realize that when you have an algorithm that works and you’re going to pitch this idea of, “Look at this wonderful thing.” Oftentimes, that algorithm has been placed on top of a curated subsection of data that is in its own enclave. That’s not connected to the real world. You show infinite profit increases. You show infinite efficiency increases in this laboratory environment. The challenge always is, how do you apply that to the real world and what do you do with it?

[bctt tweet=”100% of people just want a functioning government. We have to foster that environment and to do that, you have to get those right people working on it.” username=”talentwargroup”]

Oftentimes, what we see with people who think they want to have AI is they want integrated data. It’s pretty interesting. They go through this journey. What you find is that the algorithm needs to be applied to the real world in its raw form because the model has to change. This is where the learning comes in. There are a bunch of different ways that AI can be useful. You can apply it to supply chain things, understanding constraints, and how do you model for changes in what you do?

Our approach is about two things. To have AI or machine learning be effective at all, first, you have to have a working operating system or architecture that combines all of the real-world data into one place to then give to the algorithms. That’s a challenge. Where all these projects normally fail is when you come up with some cool algorithm and then you realize, “I have to do all the stuff that Palantir does. I have to invent that.”

You’re then like, “It’s going to take me five years to integrate all this data before I even get to the fun part of AI, applying the learning of the algorithms. Let’s see how useful it is.” Most of these projects fail because of that. Our approach is like, “How about if you skip that other part and you show up and we’re going to serve up integrated data with security and then you can lay your algorithms on top of them? You then can start the learning process, work with partners, and all this stuff.” We’re going to change how useful AI and machine learning can be to organizations.

The second part is like, “When you figure out what happens and when you get to that part where you start to get to the fun part, is this useful?” Think of computer vision. That’s this idea where you’re looking at a camera, whether this is a sensor from a satellite, a UAV, a security camera, or whatever. It’s like, “As you’re looking at this thing, are you detecting that a human being is there? Is that useful?” “I saw sheep here. I counted a car.”

There’s a low level of value in these things. What’s the useful part of this? What is unusual? How can you use algorithms to detect what is happening here that’s unusual that you wouldn’t normally find? Sometimes you get to this point where you’re looking at so much information, you’re drowning. You can’t get to the point of making a real decision.

Certain platforms have 8 to 10 sensors on them. If you have one person trying to analyze all these things and look for things, how do you get the machines to do what the machines are good at and keep humans doing things that humans are good at? Assessing, making a thoughtful decision, and understanding the broader dynamics that machines will never do. The philosophy of Palantir is to be agnostic. There’s no magic button where you press this and it says, “Find that guy.” We’ve always believed that that’s wrong.

It’s about how you collate the massive amounts of data and process it in a way that it becomes useful where the human can then be served a bunch of options, but it’s done all the hard work of filtering it first. You can always understand, at a human level, the pedigree of it. Many people are scared of AI because they think of a black box. They’re like, “I don’t know what the hell is going on here. I don’t understand it. I know that it’s saying to watch a missile or something.” We’ve always believed that’s wrong.

It’s about the idea of how can I detect a bunch of things that you wouldn’t see and bring that to the forefront? You won’t see these things because you can’t find them. It’s hidden or there’s too much information. How do you filter through that? Those are useful purposes in terms of understanding how you get to that decision.

I like to use the term glass box. How do you understand the pedigree of the pipeline that got you there? What are the data sources that it came from before it got here? What does the algorithm do in terms of its recommendations? What are the biases of recommendation and all this other stuff? You have to do that as a partnership with the organization. You’ve thought through all these things in a meaningful way where it’s keeping with the ethics of the organization.

There’s a term data lake. You don’t like the term data lake and I don’t necessarily either.

It turns into a data swamp.

It’s true. That’s why I think of it and that’s why I throw it out here. What is a data lake? It’s this lake that you throw all this information into. What are you going to do with it?

That’s hilarious. In Palantir, we’ve got a lot of business. Years ago, data lakes were all the rage. It was this idea like, “We’re going to have this new data lake. It’s going to be amazing. We’re going to take all of our disparate data.” It’s like you have 50 years of IT infrastructure and you’re like, “We have new data. We’re going to dump it in there.” You’ve integrated the data, but the problem that approach has is you lose the security, the pedigree, and the information. It’s not a live thing anymore. You’re dumping it all in there. You’re like, “I don’t know.”

A core fundamental of Palantir is that you have to keep the lineage of that data, so you understand where it came from. If you’re trying to make a real decision, you’re like, “Where does this number come from?” People start tap dancing around. Remember back in the day when the Intel officer or the S2 would be like, “Why do you know this?” That’s when the tapping starts. “It’s on the PowerPoint slide and someone else handed it to me.”

“I don’t know. I’ve been briefing on this thing since I got here.”

The idea is like, “While we’re standing here, can we audit this information? I’m going to launch a missile. Are we making a mistake here? Can we audit this? We’re making an adult decision. I want to make sure that I understand where this information comes from. There’s no messing around.” Right at that moment, the hard work that we’ve tried to solve is that this isn’t about politics and who is the biggest influencer in the room. This is objective and we know where this came from.

You mentioned the term lifetime of service to the nation. You’ve ascribed to that in all the various roles that you’ve held. You’ve also started a new venture. I want to ask you about Snowpoint Ventures. This is another opportunity where you’ve said that you are investing back in America. You’re trying to bridge this gap between the public and the private partnership and bring technology solutions to a place where the government can be much more accepting of them. What kind of ventures are you looking for? What’s the vision behind Snowpoint? What are you looking to achieve?

TJP 58 | Global Defense Lead

Doug Philippone’s lifetime of service to the nation

This goes a little bit back to your comment about the national power and our place in the World Order.

This is yet to be defined.

The idea here is America is in this strange place right now where we have to decide do we want to be the light on the hill and the actual world leader or not? China would be happy to take that place from us. When I think of the government and the Department of Defense not leaning into our best entrepreneurs or picking winners, China is taking their best stuff and they’re going all in. Our government is in this weird place where they’re competing against their own industrial base in a strange world.

When I thought about that, I said, “This might be a time when we are strangely choosing to lose.” I’m not losing to the extent that I can influence this in any way. When I thought of my journey in Palantir and what Palantir is trying to do, we did 1,000 things wrong and probably 100 things right. All of that is learning.

When I thought about this lifetime service of like, “How can I pay back or give back and foster this environment to the extent that Palantir proceeds to.” We’re just starting. It’s this idea that we won, but we’ve broken in at least. Also, to the extent that we’ve learned that you could skip these steps. How do you get new patriotic entrepreneurs who want to solve these problems for the government? How do you help them skip all those painful lessons that we had to learn at Palantir? You accelerate and do the right things and skip the bad things.

The idea is that if we do this, we’re going to get more and more entrepreneurs because they can be successful. They can solve these hard problems for the government and then get paid and it’s a real business. That’s all the goodness. I know that we will secure our borders, secure our place in the world, keep free trade, keep security, and keep democracy thriving. All this will happen, but there are the prices. We have to keep being the best.

The idea was like, “Through the venture capital firm, how can I support those entrepreneurs who want to work in this and then help them through what’s called a coalition of the competent?” How do we get this technology to the government? There’s a lot of automation in terms of an autonomous second pilot and cargo aircraft. In the military, the pilots are doing seventeen things. How do you get redundancy so you can get rid of the second pilot or the second pilot can do more things? There’s always more in the military.

How do you understand true autonomy in drones? Instead of autopilots, it’s learning. With the real augmented reality that works in planes outside, going 600 miles an hour, how can you get all that information to a pilot? How do you create real learning environments for machine learning? All these things are super important for the next level of innovation that all play on the same stuff that Palantir was doing in terms like, “I’m moving data around and making decisions.” How can I do more with less and faster?

Can I ask you about cycling?

[bctt tweet=”What are you really good at? Like extraordinarily good at? Focus on that. The best thing that you can do to actually make yourself better and well-rounded is just, don’t do those things that you suck at.” username=”talentwargroup”]

I wish I was riding my bike.

I’m going to get you there soon.

I was a cyclist when I was a kid. Strangely, it was my high school sport. It wasn’t at my high school but I was a state champion as a kid in Arizona. I loved it. I got back into it. I joined the Army and didn’t do it for a few years. When I was at West Point, I was on the cycling team. I did that. When I went back into the military, I didn’t do it forever.

Once I was retired, I had three back surgeries. Luckily, the one thing that I could do was ride a bike. During my recovery period, I got back into racing. It was great in terms of low impact. I’m super competitive. For me, it’s that outlet where you can be doing intervals and bleeding out of your eyeballs. You get rid of all the stress in the world and you’re thinking about your power output, what’s your heart rate, and setting your time. There’s a lot of strategy in it. I love cycling.

How’s the Palantir cycling team?

It’s not much of a team.

Is it an army of one? It’s you.

They don’t necessarily ride for Palantir. We have a lot of good athletes. There’s this one young guy who was the men’s elite district championship for the Mid Atlantic region, which is a big deal. He could be a domestic pro. A domestic pro cyclist makes about $12,000 a year.

He’s got to do something else.

He’ll live on his parent’s couch and get a free bike. That’s about it. Cycling has always been amazing for me. I need more time in the day.

Doug, as we close out, the Jedburghs had to do three things in World War II as core foundational tasks to be successful. They had to be able to shoot, move, and communicate. If they did these three things with the utmost proficiency, their energy and focus could be on other things, other more complex challenges, making sense of the chaos that came their way. What are the three things that you do every day as core foundational tasks to set yourself up for success in your world?

TJP 58 | Global Defense Lead

Doug Philippone’s Three Foundations to Success

I don’t know if I’m going to give you three things, but I can certainly answer a little bit of the question. If I think about a superpower, what is my superpower? We talk about that a lot at Palantir, focusing on your superpower. This is something different than military leadership. It’s about getting everybody well-rounded.

When I went to Palantir, there was something interesting that was way more about specialization. I had to learn that for a while. You’re like, “What are the incremental gains of being well-rounded?” It’s like, “When I say the real world, it’s this funny term.” I was like, “Some of the stuff is not military. It’s about specialization.”

I still call it that. I’ve been out for over six years.

Part of specialization is about understanding your superpower. A lot of folks at Palantir have gone to Harvard, Yale, and MIT. They were always the best of the best. They’ve never been told that they did anything bad in their entire life. They’ll show up to Palantir, “You suck.” To answer your question, part of this understanding is, what are you extraordinarily good at? Focus on that. The best thing that you can do to make yourself better and well-rounded is don’t do those things that you suck at.

There’s usually a lightbulb moment for people. Can you realize that or not? You’re not good at everything. Focus on the thing you’re doing. I don’t know exactly what my superpower is. One of the things that I do that I’m pretty good at is the idea of what can I do today that is going to move the needle? There’s so much noise. There’s so much like, “Does this matter?” It’s probably annoying because I blow stuff off all the time. I’m like, “That’s stupid.” I’m not doing that and maybe nobody will call me. I did that my whole life and people are used to it.

There is this important time of thinking about what can I do that will move the needle and make an extraordinary impact? Figure out what that is and focus on that. Don’t let that fall off the plate. To be clear, the losers measure work. Think about that for a second. The work doesn’t matter unless you’re talking about it retrospectively. Measuring inputs is another way. I’m not measuring inputs. I’m measuring the output.

Did you win? Let’s talk about that. How are you going to win? Figure out the decisive point. This is where military training is useful for me. It’s understanding how do you figure out how to win? What are the supporting efforts? What do I need to do to win? Focus everything on the decisive point and then what are the enablers I need to do? This is where my journey has been quite useful for me, figuring that out. I’m not saying this in the sense that I have some formula and I pull out the doctor. That’s not me.

He’s got a lot less. We love Stan. We cleared that out.

There’s intuition. You’re like, “This is good. I know how to do this.” I will see through the matrix and be like, “This is what we need to do and this is how we’re going to get there. I need to do things.”

I’ve got your three. Focus on my superpower. Focus on what moves the needle. Focus on what it takes to win.

I’ll take that.

All of our episodes are about the nine characteristics of elite performance defined by Special Operations Command. We tie them into the conversation. We talked about many of them here, drive, resiliency, adaptability, humility, integrity, curiosity, team ability, effective intelligence, and emotional strength. To be a high performer in any organization, especially as we’ve taken those from soft and we use that framework to recruit, assess, and select talent. You have to display those in varying degrees, depending on the situation that you’re in.

I say that because you certainly have exhibited all of those throughout your career and what you do today. I think about one when I sit across from my guests at the end of our conversation. This is the one that I believe you exemplify. For you, it’s curiosity. The reason I say that is because curiosity is the core, tenant, and characteristic that drives innovation. It’s what challenges the status quo. It’s drive, resiliency, and adaptability.

Those are built upon somebody’s ability to wake up every morning and say, “Can I do it better? Can I do it differently? Can I motivate myself, my team, and everyone around me to go to the next level?” when I talked to you, when I learned about you as we’ve spoken, as we sit here and have this conversation, so much of that comes down to the challenge of the status quo. Be better tomorrow than we were today. It starts with that curiosity.

Doug, thank you for having me here. I appreciate so much this conversation learning about you. I have one more question and then you got to get cycling. What does my t-shirt say? We have to have Dan get the camera on the other angle so that we can do it. You’re going to need to figure it out. I’ll give it to you. It took me a long time.

In Palantir, we have this early history of the company. We had these element t-shirts. Every time there was a new version of Palantir, we would have a new t-shirt. It started with carbon and then went through the periodic table. There are also team shirts and so forth. This is one that I don’t think I’ve ever seen before.

Are you ready?


It says, “Don’t quit my day job.” What it says here is, “Professor Cold Heart.” It took me a long time to figure it out.

I’ll do some research and figure out who made the t-shirt. That’s pretty funny. Hopefully, some humans will find this useful.

Thanks, Doug.

Thank you.


Important Links

To Top of Webpage