Meet the IIT Bombay grad who said no to Zuckerberg's $1M job offer, now building own AI startup
IIT Bombay graduate and AI researcher Rishabh Agarwal walked away from Meta despite receiving a massive compensation offer, saying it was far higher than the widely reported $1 million. He is now building Periodic Labs, an AI startup focused on accelerating scientific discoveries.
When Meta CEO Mark Zuckerberg launched an aggressive hiring push for top AI talent, reports suggested the company was offering eye-watering compensation packages worth millions of dollars. Among the researchers who received one of those offers was IIT Bombay graduate Rishabh Agarwal. But instead of staying at Meta, Agarwal decided to walk away and build his own AI startup. The story resurfaced after an X user shared a post claiming that Agarwal, who secured All India Rank (AIR) 33 in JEE and studied Computer Science at IIT Bombay, turned down a job worth around $1 million (over Rs 8.5 crore) after working at Meta for just a few months.
Agarwal himself responded to the viral post, adding an interesting detail. "Meta's offer was an order of magnitude higher than $1 million," he wrote on X, suggesting the compensation package was significantly larger than what the viral post had claimed.
Agarwal is one of the most accomplished AI researchers from India. After graduating from IIT Bombay with a Bachelor's degree in Computer Science and Engineering, he went on to complete a PhD in artificial intelligence at Mila, the Montreal-based AI research institute, under renowned researchers Aaron Courville and Marc Bellemare.
Over the years, he built an impressive resume by working at some of the biggest names in AI. His career includes stints at Google Brain, Google DeepMind, Waymo and Meta Superintelligence Labs. He has also served as an Adjunct Professor at McGill University.
According to his GitHub profile, his research focuses on reinforcement learning (RL) and large language models (LLMs). He has received a NeurIPS Outstanding Paper Award, contributed to Google's Gemma and Gemini AI models, and is credited with research that helped advance offline reinforcement learning and LLM training techniques.
Why he left Meta
Agarwal joined Meta Superintelligence Labs in April 2025 to work on reinforcement learning and reasoning for large language models. However, he resigned after about five months.
Announcing his decision on X at the time, he wrote, "This is my last week at @AIatMeta. It was a tough decision not to continue with the new Superintelligence TBD lab, especially given the talent and compute density."
Explaining his decision further, he added, "But after 7.5 years across Google Brain, DeepMind, and Meta, I felt the pull to take on a different kind of risk."
He also quoted Zuckerberg's own advice while announcing his exit: "In a world that's changing so fast, the biggest risk you can take is not taking any risk."
Building an AI startup instead
Instead of continuing at one of the world's biggest tech companies, Agarwal joined Periodic Labs as a founding member.
The startup is working on what it describes as an AI scientist capable of generating hypotheses for real-world scientific experiments. The goal is to accelerate discoveries in areas such as new materials, medicines and even physics.
Periodic Labs has reportedly received backing from Nvidia and Jeff Bezos, making it one of the closely watched AI startups in the scientific research space.
When Meta CEO Mark Zuckerberg launched an aggressive hiring push for top AI talent, reports suggested the company was offering eye-watering compensation packages worth millions of dollars. Among the researchers who received one of those offers was IIT Bombay graduate Rishabh Agarwal. But instead of staying at Meta, Agarwal decided to walk away and build his own AI startup. The story resurfaced after an X user shared a post claiming that Agarwal, who secured All India Rank (AIR) 33 in JEE and studied Computer Science at IIT Bombay, turned down a job worth around $1 million (over Rs 8.5 crore) after working at Meta for just a few months.
Agarwal himself responded to the viral post, adding an interesting detail. "Meta's offer was an order of magnitude higher than $1 million," he wrote on X, suggesting the compensation package was significantly larger than what the viral post had claimed.
Agarwal is one of the most accomplished AI researchers from India. After graduating from IIT Bombay with a Bachelor's degree in Computer Science and Engineering, he went on to complete a PhD in artificial intelligence at Mila, the Montreal-based AI research institute, under renowned researchers Aaron Courville and Marc Bellemare.
Over the years, he built an impressive resume by working at some of the biggest names in AI. His career includes stints at Google Brain, Google DeepMind, Waymo and Meta Superintelligence Labs. He has also served as an Adjunct Professor at McGill University.
According to his GitHub profile, his research focuses on reinforcement learning (RL) and large language models (LLMs). He has received a NeurIPS Outstanding Paper Award, contributed to Google's Gemma and Gemini AI models, and is credited with research that helped advance offline reinforcement learning and LLM training techniques.
Why he left Meta
Agarwal joined Meta Superintelligence Labs in April 2025 to work on reinforcement learning and reasoning for large language models. However, he resigned after about five months.
Announcing his decision on X at the time, he wrote, "This is my last week at @AIatMeta. It was a tough decision not to continue with the new Superintelligence TBD lab, especially given the talent and compute density."
Explaining his decision further, he added, "But after 7.5 years across Google Brain, DeepMind, and Meta, I felt the pull to take on a different kind of risk."
He also quoted Zuckerberg's own advice while announcing his exit: "In a world that's changing so fast, the biggest risk you can take is not taking any risk."
Building an AI startup instead
Instead of continuing at one of the world's biggest tech companies, Agarwal joined Periodic Labs as a founding member.
The startup is working on what it describes as an AI scientist capable of generating hypotheses for real-world scientific experiments. The goal is to accelerate discoveries in areas such as new materials, medicines and even physics.
Periodic Labs has reportedly received backing from Nvidia and Jeff Bezos, making it one of the closely watched AI startups in the scientific research space.
