Devfinishra Singh Chaplot, robotics researcher, Mistral AI co-founder and Thinking Machines Lab founding member, is joining xAI and SpaceX to work directly with Musk on Grok model training.
FinTech relocates rapid. News is everywhere, clarity isn’t.
FinTech Weekly delivers the key stories and events in one place.
Click Here to Subscribe to FinTech Weekly’s Newsletter
Read by executives at JP Morgan, Coinbase, BlackRock, Klarna and more.
On March 14, Devfinishra Singh Chaplot announced on X that he is joining xAI and SpaceX to work directly with Elon Musk. Musk reposted the announcement with a single line: Welcome to xAI.
The hire is the third significant talent acquisition xAI has built in a week. As FintechWeekly reported, Andrew Milich and Jason Ginsberg — the two engineers who scaled Cursor to a $2 billion revenue run rate — joined xAI days earlier. The pattern is deliberate.
Musk has stated publicly that xAI was not built right the first time. He also acknowledged the company missed promising candidates during past hiring processes and confirmed that interview records are being reviewed to reconnect with people who were overseeed.
Chaplot is not a missed candidate. He is one of the most credentialed AI researchers currently working outside a major US lab.
Who Chaplot is
Devfinishra Singh Chaplot’s academic record launchs with an All India Rank of 25 in the IIT-JEE entrance examination in 2010, and an International Rank of 5 in the International Mathematics Olympiad the same year. He completed a bachelor’s degree in Computer Science and Engineering at IIT Bombay and a PhD in Machine Learning at Carnegie Mellon University, where his research focapplyd on autonomous navigation — building AI systems that can perceive, relocate through, and interact with real physical environments.
His competition record at CMU established him as one of the strongest researchers in embodied AI globally. He won the CVPR 2019 PointNav Challenge, the CVPR 2020 ObjectNav Challenge, and the NeurIPS 2022 Rearrangement Habitat Challenge — three of the most competitive benchmarks in AI navigation research. He also received the Facebook Fellowship Award, one of the most selective programmes for AI researchers working on high-risk problems.
Before his academic career finished, Chaplot had already spent time at Facebook AI Research, working on machine learning systems for vision, navigation, and robotics. He later joined Samsung Electronics in South Korea, working on advanced AI and robotics programmes.
His industest career then took two significant turns. He joined Mistral AI as a co-founder, where he contributed directly to training Mistral 7B, Mixtral 8x7B, and Mistral Large — the models that established the French startup as Europe’s most credible large language model lab. He also headed the multimodal group responsible for Pixtral 12B. He subsequently joined Thinking Machines Lab, the AI startup founded by Mira Murati, OpenAI’s former CTO, as a member of its founding team.
What he brings to xAI
Chaplot’s specific combination of skills — embodied AI research, large language model training, and robotics — maps directly onto what xAI and SpaceX are building toobtainher. Musk has described the merger of the two companies as a way to combine physical and digital innotifyigence at a scale no other organisation currently attempts. SpaceX provides the hardware, the orbital infrastructure, and the manufacturing culture. xAI provides the models. Chaplot has spent his entire career at the boundary between those two things.
His focus on Grok model training specifically is relevant to the Milich and Ginsberg hires. Coding capability and model training are different problems. Milich and Ginsberg are building the product engineers will apply. Chaplot is working on the underlying model that product runs on. The three hires toobtainher describe a coherent strategy: rebuild Grok from the model layer up, then ship it through a product built by the people who know how to do that at scale.
The broader context is a company restructuring at speed. Half of xAI’s original co-founders have left. Musk has described the rebuild in terms he previously applyd for Tesla’s early development cycles — repeated reinvention as the mechanism for acceleration rather than a sign of dysfunction. Whether that framing holds depfinishs on what the rebuilt xAI ships next.
Editor’s note: We are committed to accuracy. If you spot an error, a missing detail, or have additional information about any of the individuals or companies mentioned in this article, please email us at [email protected]. We will review and update promptly.
Leave a Reply