The Olympiad Gold Medalists Dominating AI: Inside Silicon Valley's Most Exclusive Network

A small circle of former Hudson River Trading interns and International Olympiad gold medalists is quietly reshaping the AI landscape, with companies like Cognition, Perplexity, and Scale AI reaching valuations in the tens of billions. The group, which includes Pika co-founder and CEO Demi Guo, emerged from a shared background in elite mathematics and programming competitions during high school, then reconnected during internships at the quantitative trading firm Hudson River Trading (HRT) in the mid-2010s .

The story of this network began when Hyperliquid founder Jeff Yan participated in HRT's selective internship program during his junior year at Harvard, where only 10 interns were chosen. What made this cohort remarkable was not just their professional success, but their shared history competing in the International Olympiad in Informatics and other elite competitions. Many had already known each other since high school through these competitions before their paths crossed again at HRT .

Who Are the Key Players in This AI Network?

The group spans multiple AI companies with substantial valuations. Alexandr Wang, head of Meta AI and former founder of Scale AI, is perhaps the most prominent figure. Born in 1997 in Los Alamos, New Mexico, Wang qualified for the Mathematical Olympiad program in 2013, the U.S. Physics Team in 2014, and reached the finals of the U.S. Computer Olympiad in 2012 and 2013. He dropped out of MIT in 2015 and founded Scale AI the following year, which reached a $7.3 billion valuation by 2021 .

Other members of this network include:

  • Scott Wu: Co-founder and CEO of Cognition, an AI software engineering company valued at $10.2 billion as of September 2025, with annualized revenue reaching $400 million by early 2026
  • Johnny Ho: Co-founder of Perplexity, an AI search engine company valued at $20 billion according to the latest Forbes data, with a personal net worth of $2.1 billion
  • Jesse Zhang: Founder and CEO of Decagon, an AI startup founded by this network member
  • Demi Guo: Co-founder and CEO of Pika, a video generation AI company
  • Steven Hao: CTO of Cognition and Olympic gold medalist

These individuals share more than just professional ambition. Many won multiple gold medals at the International Olympiad in Informatics. Scott Wu, for example, won three gold medals including first place in 2014. Johnny Ho also won three gold medals, including a perfect score in 2012 .

How Did This Network Form and What Makes It Unique?

The formation of this network happened in stages. First, many members met during high school through elite mathematics and programming competitions. Then, during their college years, they reconnected at Hudson River Trading's internship program, which served as a catalyst for deeper professional relationships. The timing was crucial; these were young people entering the workforce just as artificial intelligence was beginning its explosive growth phase .

What distinguishes this group from other tech networks is their deliberate self-awareness about their potential impact. When Alexandr Wang was just 19 years old, he told his friends, "Why can't it be us?" referring to the legendary "PayPal Mafia" of the early 2000s, which included Elon Musk and Peter Thiel. A decade later, that prediction appears to be coming true .

The parallel to the PayPal Mafia is striking. Just as that earlier cohort spawned multiple billion-dollar companies and shaped the trajectory of technology for two decades, this group of Olympiad champions is now doing the same in artificial intelligence. Their companies collectively represent some of the most valuable AI startups in the world.

What Are the Major Achievements of This Network's Companies?

The commercial success of these companies demonstrates the network's impact on AI development. Cognition, founded by Scott Wu, released Devin in 2024, described as the world's first autonomous AI software engineer. This product can independently complete code writing, testing, and deployment, and significantly outperformed GPT-4 on the SWE-bench benchmark test. The company raised $175 million in May 2024 and another $400 million in September 2025 .

Perplexity, co-founded by Johnny Ho, has grown to 15 million monthly active users by April 2024 and has made several bold acquisition proposals, including a takeover bid for TikTok's U.S. operations in early 2025 and a proposal to acquire Google's Chrome browser for $34.5 billion in August 2025 .

Meta's acquisition of a 49% stake in Scale AI for $14.3 billion in June 2025 brought Alexandr Wang into Meta's leadership. He now heads Meta Superintelligence Labs (MSL), which released Muse Spark in April 2026. Muse Spark is a native multimodal inference model that supports tool calls, visual chaining, and multi-agent orchestration, representing what Meta describes as its most powerful model to date .

How to Understand the Impact of This Network on AI Development

The influence of this group extends beyond individual company valuations. Their collective approach to AI development reflects shared values and methodologies developed through years of competitive problem-solving. Here are the key ways this network is shaping the AI industry:

  • Talent Concentration: The network demonstrates how elite educational backgrounds and early competition success create lasting professional relationships that concentrate talent in high-impact companies
  • Cross-Pollination of Ideas: Members maintain relationships across different AI companies, allowing ideas and approaches to flow between organizations working on different AI problems
  • Rapid Scaling: Companies founded by network members have achieved billion-dollar valuations in remarkably short timeframes, suggesting efficient execution and access to capital
  • Ambitious Goal-Setting: The willingness to make bold acquisition proposals and pursue moonshot projects reflects a shared culture of ambition that traces back to their competitive backgrounds

The emergence of this network raises interesting questions about the role of early talent identification and elite competition in shaping technological progress. Unlike the PayPal Mafia, which formed around a specific company and then dispersed, this group maintained connections through shared competitive experiences and deliberately chose to work in adjacent fields where they could support each other's success .

As artificial intelligence continues to advance, this network of Olympiad champions and their companies will likely remain central to the industry's development. Their shared history, mutual respect, and demonstrated ability to build billion-dollar companies suggest that the next decade of AI innovation may be shaped significantly by this tight-knit group.