Sarvam Hires AI Expert Devendra Chaplot to Lead Trillion-Parameter Model Ambitions

Sarvam, an Indian AI startup, has hired Devendra Chaplot as an advisor as it pursues major expansion. Chaplot, a prominent AI researcher, brings experience from top labs including Mistral AI, Thinking Machines Lab, and Elon Musk's xAI. Sarvam announced plans to build a trillion-parameter foundation model and expand its research presence in the United States. The company recently raised $234 million in the first close of its Series B funding round, reaching a $1.5 billion valuation.

Key Highlights

• Sarvam hires Devendra Chaplot as advisor to lead trillion-parameter AI model project.
• Chaplot brings experience from Mistral AI, Thinking Machines Lab, and xAI.
• Sarvam raises $234 million at a $1.5 billion valuation for AI research expansion.
• Company plans to build AI models tailored for India's languages and cultures.

Devendra Chaplot's Background

Devendra Chaplot joined Sarvam in July as an advisor and will work from the company’s US research base in Palo Alto, California. He holds a BTech in Computer Science and Engineering from IIT Bombay. Chaplot then completed a Masters in Language Technologies and a PhD in Machine Learning at Carnegie Mellon University. His doctoral research focused on intelligent autonomous navigation, combining robotics, computer vision, and artificial intelligence.

Before finishing his PhD, Chaplot interned at Samsung Electronics and Apple. He later spent over five years at Facebook AI Research (FAIR), working on embodied AI, robotics, and computer vision. At FAIR, he contributed to systems that help machines understand and navigate the physical world.

Experience at Leading AI Labs

In 2023, Chaplot became a founding researcher at Mistral AI, a French startup challenging OpenAI in Europe. He helped develop models such as Mistral 7B, Mixtral 8x7B, Mistral Large, Pixtral 12B, and Pixtral Large. Chaplot also helped establish Mistral's US office in Palo Alto, gaining experience in scaling frontier AI startups.

In early 2025, Chaplot joined Thinking Machines Lab as Tech Lead for Data and Pre-training. He worked on Tinker, the company's training API, and contributed to its large language model infrastructure. Chaplot then worked briefly at xAI, now known as SpaceXAI, focusing on superintelligence research for about two months.

Sarvam's AI Ambitions

Sarvam aims to build India's own frontier AI models and compete with global leaders like OpenAI and Anthropic. The company is developing a foundational AI model with more than one trillion parameters. Sarvam is also expanding research in coding, speech, vision, cybersecurity, and scientific AI. Chaplot’s appointment as advisor brings expertise from leading AI labs to Sarvam’s efforts.

Chaplot emphasized the need for AI models tailored to India’s diverse languages and cultures. He stated that building models in India is essential for customization. Chaplot will work part-time from Sarvam’s US research lab as the company grows its presence in Silicon Valley.

With Chaplot on board, Sarvam strengthens its position to develop advanced AI models and expand its global research footprint.