Meta’s loss is Thinking Machines’ gain

Weiyao Wang, an eight-year veteran of Meta's multimodal perception team, has joined Thinking Machines Lab (TML), marking the latest high-profile move in a sustained talent exchange between the AI giant and the ambitious startup. The move underscores a critical trend where well-funded challengers are directly competing with established incumbents for elite researchers, a battle intensified by TML's recent infrastructure and hardware advancements.

Technical Talent and Infrastructure Firepower

TML's hiring from Meta's research divisions has been particularly aggressive. The startup now counts Soumith Chintala, co-founder of the ubiquitous PyTorch framework, as its CTO, and Piotr Dollár, co-author of the influential Segment Anything model, among its technical staff. They are part of a growing list of former Meta researchers, including specialists in multimodal models and LLM training, who have made the switch. This talent acquisition coincides with TML's major operational expansion:

• A new multibillion-dollar cloud agreement with Google.
• Access to NVIDIA’s latest GB300 chips, making it one of the first startups to use the hardware.
• Infrastructure footing comparable to industry heavyweights like Anthropic and Meta.

The talent flow is not one-way. Reports indicate Meta has hired seven of TML's founding members, and last year held talks to acquire the company. However, a review of recent hires suggests TML is currently winning a significant share of senior researchers from its larger rival.

The Strategic Calculus: Equity vs. Established Giants

While Meta offers substantial, seven-figure cash compensation packages, the appeal of a startup like TML lies in its financial trajectory. Currently valued at $12 billion, TML offers top-tier researchers a considerable equity upside that remains compelling, even when compared to the massive valuations of OpenAI and Anthropic. For many, the opportunity to build foundational systems at a rapidly scaling company—backed by state-of-the-art hardware—outweighs the stability of Big Tech. This dynamic is reshaping the AI talent market, proving that access to capital and cutting-edge GPUs can put startups on equal footing with the industry's largest players.

The confluence of massive capital, significant equity packages, and access to next-generation hardware like NVIDIA's GB300 is creating a new competitive equilibrium, allowing startups like Thinking Machines Lab to successfully raid the elite research ranks of established tech giants.