Shaun not only led the controversial investment in SpaceX in 2019 but is also one of the few investors who truly understands Musk's operational system.
While Musk simultaneously heads several disruptive companies like SpaceX and Tesla, and Jensen Huang leads NVIDIA to dominate the AI hardware market, the outside world often marvels at their "superhuman abilities" but rarely glimpses the underlying logic. Investor Shaun Maguire, with his perspectives as a PhD in mathematical physics, DARPA battlefield experience, and professional esports player, reveals the operational truths of top innovators and organizations in a deep interview—this is not only a decoding of two industry giants but also a profound analysis of talent assessment, organizational management, and investment decision-making.
As the key figure behind the controversial 2019 SpaceX investment, Shaun witnessed Musk's "20-person collective intelligence" in precise coordination and candidly admitted to his mistake of selling NVIDIA too early at a $60 billion market cap. He proposed the theory of a 15-level hierarchy in the realm of intelligence, revealing how high cognitive individuals can accurately assess talent levels, while ordinary people struggle to distinguish differences in top-tier abilities; he dismantled Musk's extreme talent mechanism of "giving them rope to hang themselves," interpreted the underestimated technical difficulties of The Boring Company, and restored the epic persuasion battle behind the SpaceX investment.
From cognitive leaps on the battlefield to teamwork lessons from professional esports, from experiences of being bullied in school to calibration experiences working with Nobel laureates, Shaun's personal journey illustrates that top achievements are never a single breakthrough but a multidimensional resonance of talent selection, capital allocation, cognitive frameworks, and organizational culture. This dialogue, rich in insights, will help you re-understand the essence of innovation and see the overlooked signals of strength and decision-making keys. Below, enjoy:
This article comes from the WeChat public account: Gao Fei's Electronic Avatar
This is a deep interview filled with information asymmetry. The interlocutor is investor Shaun Maguire, who not only led the controversial investment in SpaceX in 2019 but is also one of the few investors who truly understands Musk's operational system. Unlike other generic interviews, Shaun has a multifaceted background as a PhD in mathematical physics, DARPA deployments, and a professional esports player, using extremely precise language to analyze Musk's organizational capabilities, talent assessment systems, and why almost everyone underestimates the true strength of Musk's companies.
This is not a puff piece. Shaun candidly admits his judgment error in selling NVIDIA at a $60 billion market cap and openly shares the humiliating experiences of receiving a zero from a high school teacher and being forced to change his name by classmates. It is this unreserved authenticity that fills this conversation with rare insight density—from "how to assess someone's mathematical ability in 30 minutes as either Fields Medal level or an ordinary professor," to "why the technical difficulty of The Boring Company lies between Falcon 9 and its reusable version," every judgment has a clear underlying logic.
If you want to understand how top organizations operate, how to make correct judgments in situations of information asymmetry, and why most people can never comprehend true innovation—this podcast note recorded from the Relentless channel will provide you with an unprecedented perspective.
01 Musk's "Collective Intelligence": How 20 People Execute One Person's Will
1. Musk is not fighting alone; he is a precisely coordinated 20-person system
When most people discuss Musk, the focus is on him as an individual: working 112 hours a week, running multiple companies, a technological genius. But Shaun presents a severely overlooked dimension: "Elon the Collective."
This "collective" consists of about 20 core members who have worked with Musk for over 10 years, establishing an incredible foundation of trust. The key characteristics are:
Autonomous execution: These individuals can almost read Musk's thoughts and can act according to his ideas without needing to consult him in some cases.
Precise judgment boundaries: They know when to escalate decisions and when they can handle matters independently.
Not something that can be rushed: This kind of tacit understanding takes at least ten years to develop, constantly testing each other's limits.
"These 20 people can directly implement his will, executing with power, scale, and precision. This is an operational method that other Silicon Valley entrepreneurs simply do not possess."
2. "Give them rope to hang themselves": An extreme talent selection mechanism
Musk's talent system has two extremes:
Fast promotion channels: If you can truly execute, your promotion speed will far exceed that of almost any other organization.
Zero tolerance for failure: Mess up once, and you're basically out.
This mechanism creates a powerful self-selection effect. "You keep allowing people to rise at a pace far exceeding other organizations, and you build a strong loyalty with the most capable individuals. You give them something they can't get anywhere else."
Combined with generous compensation incentives, this system ultimately retains the top 1% in the organization—they are the true drivers of everything.
3. World-class talent detection: Engineering thinking from resumes
Shaun shared a detail: Musk interviewed a young man studying economics at university and directly judged, "You won't go into business development; you need to do mechanical engineering," because he could see that the individual had an engineering mindset. This person later became a very senior engineer.
This kind of judgment cannot be faked; it is a true superpower.
02 The "15-Level Hierarchy" of Talent Assessment: Why Most People Can't See the Differences
1. There are 15 clear levels in the realm of intelligence, but one can only "see down"
Shaun proposed an extremely sophisticated cognitive framework: in fields like mathematics, there are about 15 distinguishable levels, from "easily obtaining a top university mathematics PhD" to "a once-in-a-century mathematician."
The key insight is: this is a unidirectional transparent system.
High-level individuals can accurately assess which level someone is at within a 30-minute conversation (with an error of no more than 1-2 levels).
However, individuals more than three levels lower can hardly distinguish differences between higher levels.
For high school math teachers, all students scoring 800 on the SAT math (the maximum score) look the same—but among these students, some will become Fields Medal winners, while others are just ordinary top university graduates.
2. The analogy of chess ratings: Why a 1000-rated player cannot distinguish between 2400 and 2800
Shaun made a brilliant analogy using chess ratings (Elo Rating System):
2850 (world champion level) vs. 2700 (ordinary grandmaster): a crushing victory, with a win rate of over 99%.
2700 vs. 2500 (just became a grandmaster): a clear advantage, with a win rate of 85-90%.
But if you let a 1000-rated player look at the game records of these matches, they cannot distinguish the level differences at all.
Conversely, a 2800-rated player can accurately assess the opponent's level after seeing just 10 moves.
"This is a core ability for investors: my main job is to assess people and talent, especially in the earliest stages. Understanding the level differences in talent is an absolute superpower."
3. The source of calibration ability: Personal exposure to truly top talent
Shaun's calibration ability comes from multiple close encounters with extreme talent:
In mathematics: In 2010, he participated in a summer program at the Clay Mathematics Institute in Brazil, spending a month with 100 top mathematicians, including 2 Fields Medal winners and 1 awardee that year. "We 30 graduate students spent all day together, competing with American and French students. The closeness of the French mathematics community is unimaginable; they almost all come from two universities (École Normale Supérieure and Polytechnique). They are fiercely competitive, completely different from the American stereotype of mathematicians."
In algorithmic trading: During an internship at DRW, a colleague was a three-time Putnam Competition champion and a three-time International Mathematical Olympiad gold medalist, who later became a tenured professor at MIT.
In physics: He knows several Nobel laureates (before they won), such as Kip Thorne.
The value of these experiences is: when you witness what 0.001% of people are like, your standards for judgment are forever changed.
03 The Underlying Methodology of Investment Decision-Making
1. The first question is always: What capabilities does this company need?
Shaun's investment framework is very clear:
First assess: What traits are necessary for this company to succeed? Some companies do not require intelligence (like a regular garbage collection company), but if it's a "garbage collection robot," intelligence becomes crucial.
Then evaluate: What level is the founder at in these most important traits?
"It could be sales ability, it could be raw mathematical ability (for an AI research company), or it could purely be stress resistance. The key is to first clarify what traits are truly important, and then assess the individual's level in that dimension."
2. Case study: How to assess 2600+ level technical ability from a cold email
Shaun invested in an AI code generation company, Factory, where the founder Matan Grinberg impressed him through a cold email:
"This founder mentioned in the email that he co-authored a paper with Juan Maldacena during his undergraduate studies. Juan is one of the most renowned string theorists in the world. For me, just this information means—this person's technical ability is at least at the 2600 level (using chess rating as a comparison)."
Key insights:
If it was during graduate studies collaborating with Juan, the threshold would be around 2300.
But to collaborate and publish a paper with Juan during undergraduate studies means 2600+.
"He only said 'co-authored a paper with Juan' instead of 'Juan Maldacena'—I appreciate this high-level way of communication."
More importantly, this founder has both strong technical ability and excellent sales skills and empathy—this is the true magic combination.
3. Why few VCs can capture these signals
Shaun candidly states: "Most VCs would not recognize any signals from a cold email that says 'co-authored a paper with Juan.' Because they lack the calibration ability in this field."
This also explains why truly good projects are often missed by "outsiders"—it's not that the founders aren't good enough, but that the evaluators simply don't understand.
04 The 2019 SpaceX Investment: An Epic Persuasion Battle
1. Why Investing in SpaceX at the Time Was "Utterly Crazy"
When investing in SpaceX in 2019, Tesla's market cap was only $4-5 billion, and Musk did not yet have the trillion-dollar companies as endorsements that he has today. The investment discussion was the "most controversial and intense conversation" Shaun had ever experienced—one partner even gave it a score of 1 out of 10.
But Shaun did not accept "no" as an answer.
2. Persuasion Strategy: Starting with a $20 Million Small Bet
Faced with significant resistance, Shaun adopted a gradual strategy:
Step One: A month of persistent pursuit, ultimately securing a $20 million "test investment" (the team was initially seeking $600 million).
Step Two: Over the next six months, sending progress updates to all decision-makers every three weeks.
"This had two effects: first, it showed them your persistence; this is not a fleeting idea; second, it allowed them to see longitudinal data—seeing the speed and acceleration of progress. You can often change people's minds this way."
This methodology applies to all scenarios, including information warfare: it's hard to change people's minds with a single data point, but over time, a continuous flow of data can recalibrate perceptions.
3. Why Today's Consensus Was "Non-Consensus and Not Obvious" at the Time
Shaun emphasized that investing in SpaceX today seems like an obvious good deal, but in 2019:
Starlink had not yet been proven feasible.
Reusable rockets were just starting to work.
Tesla was not yet a trillion-dollar company.
The entire aerospace industry was not understood by mainstream investors.
The success was not due to luck, but because he had the ability to see the level differences that others could not.
05 Why Almost Everyone Underestimates Musk's Companies
1. The Boring Company: Misunderstood Technical Difficulty
Shaun asked Steve Davis (a former SpaceX employee and now head of The Boring Company): What is the difficulty of the "Zero People In Tunnel, Continuous Mining" boring machine within the SpaceX technology spectrum?
Steve said: It's slightly harder than Falcon 9, but easier than the reusable version of Falcon 9.
This is a shocking judgment: the engineering difficulty of this equipment is higher than that of Falcon 9—but the outside world completely fails to understand this.
The reason is simple: people only make linear comparisons. They compare one boring machine to another but fail to see the generational differences between Falcon 1, Falcon 9, the reusable version of Falcon 9, and Starship.
2. "Precedents and Superlatives": The Cognitive Limitations of the Human Brain
Musk has a saying: people only respond to "precedents and superlatives."
Precedents: This means you have broken through the nonlinear stage and truly achieved the goal.
Superlatives: This refers to visual impacts like rocket explosions or landings—differences that ordinary people can intuitively understand.
But people cannot comprehend:
The differences in payload capacity.
The differences in achievable orbits.
The differences in underlying technical complexity.
"So The Boring Company is now like SpaceX before 2009—before reaching the milestone of 'Zero People In Tunnel, Continuous Mining,' the outside world cannot perceive progress. But once achieved, the perception will jump tenfold."
3. Optimus Robot Demonstration: Deliberately Created "Moment of Extremity"
Shaun saw 20 Optimus robots walk out at a Tesla event:
"They walked over from 30-40 feet away, and at first, I really couldn't tell if they were human actors or robots. I first looked at their faces—just like looking at people. Then I started looking down at their bodies until I saw their waists—very narrow, not like humans—only then did I confirm they were real robots."
This experience created a nonlinear psychological impact, making people truly feel that "the future is coming." This is what Musk excels at: creating milestone moments that are shocking, allowing people to intuitively understand what is about to happen—something that charts and data can never achieve.
06 The Art of Capital Allocation: The Wisdom of Bet Sizing
1. Musk is "One of the Greatest Capital Allocators of All Time"
Shaun believes Musk's capital allocation ability is underestimated:
The timing of battery factory investments.
The timing of autonomous driving investments (the actual effective time points).
The timing of the Starlink low Earth orbit internet constellation.
The key is not "what was invested," but "when and how much was invested."
2. The Evolution of the Starlink Bet: The Subtle Rhythm from Small to Large Bets
SpaceX began researching Starlink around 2013, but the investment was small—they were waiting for Falcon 9 to achieve reusability. After achieving this in 2016, it took another two years to resolve technical details. By around 2018, Musk confirmed that all elements were in place:
Phased array RF technology was sufficiently mature.
Launch costs were economically viable (because rockets were reusable).
The number of flights per year could be significantly increased (not needing to build a new rocket each time).
"At this point, the small bet upgraded to a medium bet. Once the unit economic model of the medium bet was validated, it then upgraded to a large bet."
3. The Hedge Fund Manager's Mindset: Learning with Small Positions, Then Gradually Increasing Exposure
Shaun used a hedge fund analogy:
"If you have a vague idea about a company but don't fully understand it, you would first buy a 1% position—this allows you to start truly learning about the company. It's hard to learn deeply from the outside, but with a position, you will have emotional investment and pay more attention. If the thesis is validated, then increase from 1% to 5% or 10%."
Musk's intuition in this area is severely underestimated—his judgment on bet sizing is as precise as his judgment on technology.
07 From the Battlefield to Investment: Cognitive Leaps in Extreme Environments
1. The Most Extreme "Stepping into the Fire" Experience
During his PhD, Shaun was recruited by DARPA (the U.S. Defense Advanced Research Projects Agency) to work in Afghanistan. This was completely outside his plans (he originally wanted to be a professor), but he said: "I want to serve my country; this is the greatest way to step into the fire."
"In a war zone, you learn at an incredible speed—because of the level of feedback and danger."
2. A Coordinated Attack Incident: When Intuition Outpaces Data
In March or April 2012, Shaun had to drive from one base to another every Friday to report. That day, there were no cars on the road—what usually took 45 minutes to an hour took only 15 minutes.
"I immediately felt something was off. Upon arriving at the base, I went straight to the intelligence briefing center and asked everyone: What’s going on? What’s in the intelligence report? Everyone could feel the atmosphere was wrong, but there was no clear intelligence indicating what would happen. About 2-3 hours later, a large-scale coordinated attack occurred—six different bases across the country were attacked simultaneously."
This experience made Shaun deeply realize: when your intuition tells you something is wrong, but you can't find it in the data—that indicates the data system needs improvement.
He spent three weeks thinking about "how we could possibly know in advance"—the answer is that there are many ways, but he wouldn't specify what they are in the interview.
08 From Being Bullied to World Champion: The Power of Non-Traditional Paths
1. The Absurd Experience of Being Renamed "Joel" on the Second Day of 7th Grade
On the second day of 7th grade, a wealthy student named Vinnie Terramina (whose father sold a garbage company to Waste Management for $500 million) stood on a flower bed and pointed at Shaun in the crowd, saying: "You, your new name is Joel."
As a result, on every teacher's roll call, "Sean Maguire" was crossed out and replaced with "Joel." Other students had told the teachers in advance, "This Sean is a troublemaker; his real name is Joel, but he will mess with you." The whole class laughed, and the teacher was confused.
This name followed him throughout high school. Some people didn't even know his real name was Sean. Years later, Shaun asked Vinnie why, and he replied: "I was hungover that day; you looked like a kid I knew named Joel Jacobson. It just came out. Sorry."
2. The "Lord of the Flies" School Where Sleeves Were Torn Off
The school had a tradition in gym class: some students would tear the sleeves off their uniform T-shirts, and then more students would follow suit, eventually turning into "tearing off others' sleeves" to display dominance.
Shaun and another fast kid, Chris Ringstrom, were chased by over 20 students every day for 15 minutes just to avoid having their sleeves torn off—"we were the last two with intact sleeves."
"We would run into a flock of seagulls, hundreds of them taking off. When they took off, they would poop (because they were digesting), about 5% of the seagulls would defecate while taking off. Almost every morning, one or two of the kids chasing us would get hit by bird droppings."
This experience explains why Shaun became a professional esports player: he would rather spend 10 hours a day in Counter-Strike than be at school.
3. From Professional Esports to Investor: Key Skills Learned from Gaming
During grades 8-10 (ages 13-16), Shaun played 10 hours of CS every day, joined a top North American team, and made some money (about $10,000 a year).
What did he learn?
Basic Networking Knowledge: To reduce latency by 10 milliseconds, he learned about computer networking and used Wireshark (a network analysis tool) to analyze home network jitter.
Extreme Team Collaboration: Tactical coordination in 5v5 matches, achieving communication at a professional athlete level.
Pixel-Level Precision in Angle Understanding: For example, on the de_Aztec map, you need two teammates to stand at a 90-degree angle, so even if the opponent reacts faster and kills the first person, the second person can ensure they kill the opponent—the worst-case scenario is a 1-for-1 trade, not losing two people.
"Parents think that kids being obsessed with games is a bad thing; my view is: being obsessed with almost anything is better than indifference."
He got an F in high school Algebra 2 but eventually earned a PhD in Mathematical Physics—because gaming taught him systematic thinking and teamwork.
09 How Key Figures Work with Musk
1. The Working Styles of Gwynne Shotwell and Steve Davis
Shaun summarized the traits of those who can work with Musk for a long time:
First, a willingness to truly work hard. Antonio Gracias (a partner at Valor Equity) slept in the Tesla factory to participate in production ramp-up—"the level of commitment is hard to explain how much it deserves respect."
Second, discretion. "This sounds simple, but when you are trying to truly change the future, sometimes you need to be unexpected. Most investors leak information. Keeping a secret is a very low bar, but most people fail to do it."
Third, all-weather support. Being present in both good times and bad—Antonio excels in this aspect more than almost any investor.
2. Understanding "AlphaGo-style Decisions": Moves That Can Only Be Understood After 17 Steps
Shaun compared Musk to AlphaGo (the AI that defeated the world champion in Go):
"AlphaGo makes moves that some Go players completely fail to understand, but after 17 moves, AlphaGo wins, and everyone realizes, 'Wow, that move was insane.' Working with Musk is like that—he makes some extremely counterintuitive moves that you can't understand in the moment, but after 6 months or a year, you get it, and it looks prophetic."
Those who do not understand this will leak plans and question decisions. Those who do understand will believe in the system, trust the direction, be loyal to teammates, and be mission-oriented—these traits, combined with capability and the courage to bet on oneself, will lead to great success.
3. A Culture Where "Anyone Can Raise Their Hand to Take on Tasks"
In Musk's organization, anyone can raise their hand and say, "I'll do this."
If you really accomplish it, great, you will be promoted.
If you raise your hand again and complete it, you will be promoted again.
But if you say you can do it and fail—you’re done, or you’ll be stuck in that position.
The key is knowing your limits. It is perfectly acceptable to admit, "I'm not sure I can do this" or "I think someone else is more suitable"—this is different from proactively taking on tasks. Being assigned a task and voluntarily taking on a task have different requirements.
10 Two Major Investment Mistakes and Reflections
1. Selling NVIDIA for $60 Billion: Underestimating Jensen Huang
Shaun bought NVIDIA from its IPO in 1999 (when he was 13, due to gaming graphics cards) and held it until it reached a $600 billion market cap. At that time, data center revenue and gaming revenue were roughly equal, and he thought the valuation was too crazy, so he sold.
Where did he go wrong?
First, underestimating Jensen Huang. "Even though I held this company for a long time, I didn't realize how incredible Jensen Huang is. He really excels at maximizing his advantages in an unbelievable way."
Second, underestimating how market irrationality can become a self-fulfilling prophecy. Shaun believed the market was somewhat irrational at the time, but Huang played that irrationality so well that it made NVIDIA extremely valuable, while AMD and Intel became very cheap, forcing them to cut investments. Thus, NVIDIA could ramp up investments dramatically, while AMD and Intel's catch-up plans were essentially suppressed.
Third, being driven away by "funds that don't understand hardware." Shaun was frustrated: "I have studied Broadcom, TSMC, ASML since I was young and know semiconductors inside out. But the funds that flooded in two years ago didn't even know what Mellanox (the key networking company acquired by NVIDIA) was, yet they said NVIDIA would reach a $3 trillion market cap. I thought: forget it, I'm out."
Key addition: Shaun believes NVIDIA's acquisition of Mellanox in 2019 (for about $8 billion) "might be one of the best acquisitions in history," providing NVIDIA with a significant interconnect technology moat—this is a seismic advantage in the data center era.
2. Reflection: Some Things Can Only Be Calibrated in Hindsight
Shaun believes that if he had understood how incredible Jensen Huang was at the time, this mistake could have been avoided. But he also admits that many of the funds flowing into NVIDIA knew nothing about semiconductors—sometimes, ignorance frees you from the constraints of "rational valuation."
"This is the paradox of investing: deep expertise can sometimes make you overly cautious."
He also said, "I would never short NVIDIA because I fear that irrational power, and I don't know how big the AI infrastructure scale will be."
Core Insights Q&A
Q1: Why can Musk operate multiple companies simultaneously while other entrepreneurs cannot?
A: Musk is not one person—he is a "collective intelligence" system of 20 people. These 20 people have worked with him for over 10 years, establishing deep trust, enabling them to execute his will autonomously, knowing when to report and when to make independent decisions. This kind of tacit understanding cannot be rushed; it takes at least a decade to build. Coupled with an extreme talent selection mechanism of "giving them enough rope to hang themselves" (perform well and get promoted quickly, mess up once and you're out), the result is the top 1% of the organization—this is the true core competency. Most Silicon Valley entrepreneurs do not operate this way.
Q2: How to assess a person's true ability, especially in situations of information asymmetry?
A: The key is having "calibration ability"—establishing precise perceptions of different levels by personally interacting with true extreme talents (Fields Medal mathematicians, Nobel Prize winners, world champion esports players, etc.). There are 15 clear levels in the realm of intelligence, but this is unidirectionally transparent: high-level individuals can accurately assess another's level within 30 minutes (with a 1-2 level margin of error), but those below level 3 cannot distinguish higher-level differences at all. It's like asking a 1000-rated chess player to watch a game between 2600 and 2800-rated players; they cannot see the difference—but a 2800-rated player can accurately assess the opponent's level after just 10 moves. Investing is about recognizing these signals: for example, what does "collaborating with top scholars to publish papers during undergraduate studies" mean? Most VCs cannot understand this signal at all.
Q3: Why does almost everyone underestimate Musk's companies?
A: Because the human brain only reacts to "precedents and superlatives" and cannot comprehend nonlinear progress. People can understand rocket explosions or landings, but they cannot see the generational differences between Falcon 1, Falcon 9, the reusable version of Falcon 9, and Starship. The Boring Company's goal of "Zero People In Tunnel, Continuous Mining" has a technical difficulty between Falcon 9 and the reusable version of Falcon 9—but the outside world completely fails to understand this because they only make linear comparisons (comparing one boring machine to another). The deeper reason is that people cannot understand Musk's "AlphaGo-style decisions"—those actions that are completely incomprehensible in the moment will reveal prophetic accuracy 6 months or a year later. Those who can work with Musk long-term possess the ability to "believe in the system and wait for validation"—this is extremely rare in an era that pursues immediate feedback.
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
