Andrew Tulloch, co-founder of Thinking Machines Lab, the AI startup founded by former OpenAI CTO Mira Murati, has left the company to join Meta. Tulloch’s departure comes amid Meta’s high-profile push to poach elite AI talent and accelerate its superintelligence ambitions, especially following the lackluster reception of its Llama 4 model.
Thinking Machines Lab, founded in late 2024 quickly made waves for its focus on fine-tuning open AI models. Tulloch’s move is seen as a key development in Silicon Valley’s escalating battle for top researchers.
Meta CEO Mark Zuckerberg had reportedly first attempted to acquire Thinking Machines Lab outright, a bid rebuffed by Murati. When that failed, Zuckerberg extended offers to more than a dozen of the startup’s staff, with Tulloch ultimately accepting an invitation to transition to Meta.
Sources at the Wall Street Journal suggest Tulloch’s compensation package could have reached $1.5 billion over six years, featuring rich bonuses and an aggressive stock grant, though Meta’s spokesperson dismissed the figures as “inaccurate and ridiculous.”
Tulloch brings deep technical expertise from prior stints at OpenAI and Facebook’s AI Research Group, and his return to Meta marks both a reunion and a pivotal recruitment in the company’s quest to close the gap with rivals in generative AI.
Meta’s Aggressive Push
Meta has ramped up its strategy to secure AI leadership, offering the most lucrative deals in Silicon Valley and strategically partnering with startups for tech and talent. Recent reporting reveals Meta dangling $100 million-plus signing and retention bonuses for top scientists, including OpenAI employees, as it seeks to bolster its AI divisions and execute against ambitious plans for its upcoming Superintelligence Labs.
Tulloch’s hiring, alongside others from Thinking Machines Lab emphasizes Meta’s intent to shore up expertise that can catalyze breakthroughs after setbacks with previous releases.
The AI industry has responded with fascination and debate. Reddit threads on Tulloch’s move brim with commentary about the astronomical sums offered, the challenges of developing AGI, and the broader implications for competition among the tech giants.
Many see Tulloch’s choice as emblematic of the current moment, where value is placed on research leadership and future potential in artificial intelligence development.
Thinking Machines Lab, reported that Tulloch’s decision is “for personal reasons,” and the company says it remains focused on advancing its platform for open model fine-tuning despite the loss. Tulloch’s extensive background, including distinguished engineering roles at OpenAI and Meta, and top academic achievement from Cambridge and UC Berkeley, positions him as a key player in shaping next-generation AI infrastructure.
