The AI industry is witnessing a dramatic talent shift, with Thinking Machines Lab (TML) aggressively hiring top researchers from Meta, turning the tables on a company that once tried to acquire it. This trend, highlighted by recent high-profile moves, signals a new phase in the battle for AI expertise.
Weiyao Wang, who spent eight years at Meta helping build multimodal perception systems and contribute to open-world segmentation projects including SAM3D, recently joined TML. His move comes as TML expands rapidly, having just signed a multibillion-dollar cloud deal with Google that gives the startup access to Nvidia's latest GB300 chips. This partnership, announced at Google Cloud Next on Tuesday, follows an earlier agreement with Nvidia and puts TML in the same infrastructure tier as Anthropic and Meta.
Meta reportedly held talks to acquire TML around this time last year. Since then, Meta has been picking off TML's founders one by one, while TML has been recruiting Meta researchers in return. Based on LinkedIn profiles, TML has hired more researchers from Meta than from any other single employer. Key hires include Soumith Chintala, TML's CTO who spent 11 years at Meta and co-founded PyTorch; Piotr Dollár, an 11-year Meta veteran and research director; Andrea Madotto, a research scientist in Meta's FAIR division; James Sun, a software engineer with nearly nine years at Meta; Weiyao Wang; and Kenneth Li, a Harvard PhD who spent ten months at Meta.
TML has drawn talent from beyond Meta as well, including Neal Wu, a three-time gold medalist at the International Olympiad in Informatics and a founding member of Cognition; Jeffrey Tao via Waymo, Windsurf, and OpenAI; Muhammad Maaz from Anthropic; Erik Wijmans from Apple; and Liliang Ren from Microsoft's AI Superintelligence team. The startup's headcount now stands at around 140.
Financial incentives drive talent decisions, with TML currently valued at $12 billion—a figure unimaginable for a company at this stage in any previous tech cycle. TML has released just one product so far, but compared with OpenAI and Anthropic's record-breaking valuations, there is still significant financial upside.
This talent transfer has broader implications for the AI industry, showing that smaller, well-funded startups can compete with tech giants for top researchers. It also highlights the importance of infrastructure partnerships, like TML's Google Cloud deal. The timing is critical as AI research advances rapidly, with companies needing the best talent to stay competitive.
Meanwhile, investors are closely watching Meta and Microsoft, which report earnings on Wednesday, April 29. Meta's Q1 2026 revenue is expected at about $55.5 billion, with earnings of $6.65 a share. Analysts maintain a Strong Buy consensus with an average target of $855.60. Microsoft is expected to report earnings of about $4.05 a share on revenue of roughly $81.4 billion. The market is no longer rewarding AI spending on faith alone, seeking proof that hyperscaler capex is translating into real revenue and stronger margins.
