Trapit Bansal, an alumnus of IIT-Kanpur and former researcher at OpenAI, has joined Meta’s newly established Superintelligence Labs. Bansal will focus on areas such as deep reinforcement learning, reasoning, and natural language processing, advancing Mark Zuckerberg’s ambitious efforts in artificial general intelligence (AGI).

Bansal’s professional path stands out for combining rigorous academic training with experience at leading tech firms. After earning his degree in mathematics and statistics from IIT Kanpur, he completed a PhD in computer science at the University of Massachusetts Amherst, specializing in meta-learning and natural language processing (NLP).

Bansal’s career kicked off in 2012 as an analyst at Accenture Management Consulting in Gurugram. He then delved into research at the Indian Institute of Science (IISc) in Bangalore, where he focused on Bayesian modeling and inference. Over time, he gained practical experience through internships at tech giants like Facebook, Google, Microsoft, and OpenAI, where he contributed to projects in NLP, knowledge graphs, and reinforcement learning.

In 2022, Bansal joined OpenAI as a full-time researcher, working directly with co-founder Ilya Sutskever on reinforcement learning for reasoning. His contributions were key to developing OpenAI’s first AI reasoning model, Q1, which became a critical foundation for the reasoning abilities found in ChatGPT.

At Meta Superintelligence Labs (MSL)—a division created by Zuckerberg to accelerate AGI development—Bansal is expected to play a major role in building next-generation large language models. Under the leadership of former Scale AI CEO Alexander Wang and former GitHub CEO Nat Friedman, MSL is aggressively hiring top AI researchers from OpenAI, Anthropic, and DeepMind. According to reports from Bloomberg and The Wall Street Journal, Meta has been offering compensation packages worth $300 million over four years, including stock options and massive signing bonuses. While the specifics of Bansal’s deal remain confidential, industry speculation suggests Meta offered a similarly lucrative package.

This aggressive recruitment drive has sparked tensions in the AI community. OpenAI CEO Sam Altman recently commented on Meta’s offers of “$100 million signing bonuses,” asserting that OpenAI’s top talent had declined those offers—though Bansal’s departure and others’ exits indicate Meta’s strategy is proving effective.

As Meta gears up to launch a reasoning-focused AI model, Bansal’s arrival is seen as crucial to filling the missing piece in its AGI ambitions. His deep expertise in building systems capable of reasoning, learning, and planning is expected to significantly boost Meta’s efforts to create AGI and compete with OpenAI’s GPT-4o and DeepSeek’s R1.

Meanwhile, despite Meta spokesperson Andy Stone’s efforts to downplay reports of sky-high compensation offers, industry insiders confirm that Meta’s aggressive hiring spree is reshaping the competitive landscape in AI research.