Thinking Machines Lab, the artificial intelligence startup founded by former OpenAI chief technology officer Mira Murati, has recorded a major shake up at the top as two of its co founders and another senior researcher return to OpenAI. The developments, disclosed on Wednesday, have drawn attention across the AI community, given the startup’s high profile launch and deep ties to OpenAI.
Murati announced on social media that the company had parted ways with its co founder and chief technology officer, Barret Zoph, adding that Soumith Chintala would assume the CTO role. She described Chintala as a seasoned AI leader with more than a decade of contributions to the field and a key figure within the Thinking Machines team. Her statement, however, made no reference to other departures.
Less than an hour later, OpenAI’s head of applications, Fidji Simo, revealed that Zoph would be returning to OpenAI alongside Luke Metz and Sam Schoenholz, both of whom had also been working at Thinking Machines. Metz is a co founder of the startup and a former OpenAI technical staff member, while Schoenholz previously worked at OpenAI and was still listed as part of Thinking Machines on LinkedIn.
The exits are particularly striking given the startup’s recent origins and strong financial backing. Murati left OpenAI in September 2024 to co found Thinking Machines with Zoph and Metz, positioning the company as a major new player in advanced AI research. In July last year, the startup raised a 2 billion dollar seed round led by Andreessen Horowitz, with backing from Accel, Nvidia, AMD and Jane Street, valuing the company at 12 billion dollars.
While talent movement between major AI firms is common in Silicon Valley, the loss of multiple co founders within a year of launch is unusual and potentially destabilising. Thinking Machines has already seen other senior departures, including co founder Andrew Tulloch, who joined Meta in October. The latest developments underscore the intense competition for elite AI talent and highlight the internal pressures facing even the most well funded new entrants in the rapidly evolving AI sector.