“`html

Table of Contents

IIT Bombay Graduate Leaves Meta Million Dollar Job in 5 Months

Date Published: 27 August 2025
Author: Swastika Das Sharma

TL;DR

Rishabh Agarwal, a prominent AI researcher and IIT Bombay graduate, joined Meta’s SuperIntelligence Labs for a reported million-dollar salary but resigned within five months.
He previously worked at Google Brain and DeepMind, playing a key role in reinforcement learning and large language models.
His departure highlights the volatility and fierce talent war in the AI industry between giants like Meta, Google, and OpenAI.

Introduction: The Surprising Exit That Shook Silicon Valley

The world of artificial intelligence research is a competitive landscape full of high stakes, innovation, and, increasingly, talent wars between tech giants. In 2025, one story that has caught the attention of both the academic and tech communities is that of Rishabh Agarwal, an Indian computer scientist whose whirlwind journey from Google DeepMind to Meta—and his abrupt exit after just five months—has become the talk of the technology sector.

But what made this IIT Bombay graduate, recruited by none other than Mark Zuckerberg, walk away from a coveted million-dollar role at Meta’s SuperIntelligence Labs? Here’s the full story behind this high-profile transition, what it reveals about the AI arms race, and insights into his remarkable career.

Who is Rishabh Agarwal?

Rishabh Agarwal is a distinguished AI researcher whose expertise has been honed in some of the world’s best research labs. Hailing from India and an alumnus of IIT Bombay, Agarwal’s journey demonstrates the ever-increasing importance of global talent in shaping the future of artificial intelligence.

Educational Milestones

Bachelor’s from IIT Bombay in Computer Science and Engineering
Undergraduate research on “approximating the evaluation function in Scrabble”
PhD at Mila – Quebec Artificial Intelligence Institute (Canada)

Professional Journey

Research intern at Saavn (Search and Algorithms division)
Intern at Tower Research Capital’s Algorithm Trading team
Senior Research Scientist at Google Brain (2018–2023): Worked on deep reinforcement learning (Deep RL)
Key member at Google DeepMind: Focused on large language models, reinforcement learning, self-improvement, and distillation
Adjunct Professor at McGill University, Canada

The Meta Move: Million Dollar Salary and the AI Talent Race

In early 2025, Meta, led by CEO Mark Zuckerberg, was aggressively expanding its SuperIntelligence Labs—a team dedicated to achieving breakthroughs in artificial general intelligence (AGI). Amidst the industry-wide AI race, Meta’s recruitment spree targeted some of Google DeepMind’s most talented researchers, including Rishabh Agarwal.

Agarwal reportedly received a million dollar salary to join Meta—an offer indicative of both his value and the scale of the AI arms race.
He was tasked with advancing post-training for “thinking” AI models, a key frontier for next-generation intelligent systems.

Why Did Meta Poach Google DeepMind Employees?

Fierce Competition: Tech giants are racing to build state-of-the-art AI, often recruiting aggressively from competitors.
AI Talent Shortage: The pool of experienced AI researchers is small, so companies offer lucrative packages to attract top minds.
Industry Expectations: Achieving AGI would be a landmark event, granting the leading company an unprecedented edge in tech and business.

A Sudden Departure: Why Did Rishabh Agarwal Quit Meta So Quickly?

On August 25, 2025, just five months after joining Meta, Agarwal announced—via a heartfelt post on X (formerly Twitter)—that he was leaving the company. This abrupt transition fueled speculation: What would make a respected researcher leave such an attractive opportunity?

“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. But after 7.5 years across Google Brain, DeepMind, and Meta, I felt the pull to take on a different kind of risk.”

Agarwal’s reflection hints at a desire for new challenges and personal growth beyond the corporate race for AGI—a sentiment that resonates with many in the fast-evolving tech world.

Key Factors Behind the Exit

Pursuit of “a different kind of risk”: Indicates a search for personal or entrepreneurial endeavors beyond traditional big-tech research.
Burnout and Intensity: The relentless pace and pressure in AI research can lead to quick turnover at top firms.
Alignment of Values: Some top researchers are choosing projects with direct societal benefit or academic freedom rather than big corporate ambitions.

What Did Agarwal Achieve at Meta?

Despite the short tenure, Agarwal and his team at Meta pushed “the frontier on post-training for ‘thinking’ models.” This refers to advanced work on enabling AI systems to generalize, reason, and improve via new forms of learning—an area critical for AGI development.

Progress in post-training means making large language models (LLMs) more capable after their initial development, aligning with goals of “superintelligence.”
Meta’s SuperIntelligence Labs, with talent like Agarwal, aimed to rival efforts at OpenAI, Google DeepMind, and Anthropic.

What Does This Mean for the AI Industry?

Agarwal’s short but impactful journey at Meta—and his reasons for leaving—tell us several important things about the current AI industry dynamic:

The AI Talent War is Getting More Intense

Companies are willing to offer seven-figure salaries and dream projects to secure top talent.
The exodus of researchers between firms can disrupt product timelines and shift momentum in the AI race.

Personal Values and Work-Life Balance Matter—even at the Top

Even with the most lucrative offers and “compute density” (cutting-edge resources), researchers must weigh personal satisfaction, sense of purpose, and long-term career goals.
Tweets and posts from Agarwal and peers suggest that the thrill of working at the frontier is balanced by the opportunity to pursue “a different kind of risk”—often outside the confines of Big Tech.

Impact on Meta and the Industry

Meta’s talent coup was partly undercut by quick departures—not only Agarwal but other new hires also left in similar timeframes (“Meta AI hit by talent exodus? Three staff quit within months of joining”).
Google DeepMind’s CEO and other tech leaders have called out Meta’s “poaching spree,” highlighting the high-stakes competition for a limited pool of AI experts.
OpenAI, another major player, reportedly even paused some operations to stem Meta’s talent raid.

Rishabh Agarwal’s Career: A Timeline

2018–2023: Senior Research Scientist at Google Brain, working on deep reinforcement learning.
2023–2025: Staff at Google DeepMind, contributing to advancements in LLMs, RL, and self-improvement techniques in AI systems.
April 2025: Joined Meta SuperIntelligence Labs at a million-dollar annual salary.
August 2025: Resigned from Meta after five months—his stated motivation was to seek new risks and challenges outside of corporate research labs.
Academic Side: Adjunct Professor at McGill University, indicating continued ties to academia.

Lessons for Aspiring AI Professionals

Agarwal’s journey is instructive for new graduates, researchers, and anyone eyeing a career at the intersection of AI, academia, and corporate innovation:

Continuous Learning: Deep specialization, such as expertise in reinforcement learning or LLMs, remains highly valuable.
Flexibility: Being open to new opportunities—even those outside one’s comfort zone—can pay off, both in terms of learning and compensation.
Introspection: Career decisions in today’s world aren’t just about money or prestige—they’re about purpose, alignment, and personal growth.

What’s Next for Rishabh Agarwal?

While he hasn’t publicly revealed his next steps as of writing, Agarwal’s words (“the biggest risk you can take is not taking any risk”) hint at entrepreneurial ventures or independent research—possibly with greater freedom than offered by mega-corporations.

The Big Picture: AI, Risk, and the Human Factor

The high-profile moves of AI researchers like Agarwal show us how the story of AI isn’t just about algorithms and compute power—it’s about people. The field attracts not only scientific brilliance but also those willing to question, challenge, and occasionally step away from enormous corporate offers to pursue new risks, ideas, or just a different life balance.

Whether you are passionate about technology trends, exploring a career switch into AI, or simply fascinated by how innovation happens, Rishabh Agarwal’s journey is a reminder that the future is shaped as much by human aspirations as by technological breakthroughs.

Frequently Asked Questions (FAQs)

1. Who is Rishabh Agarwal and why is his departure from Meta significant?

Rishabh Agarwal is an IIT Bombay alumnus and renowned AI researcher who left Meta just five months after being recruited with a million-dollar package from Google DeepMind. His sudden exit highlights both the volatility of the AI talent market and the personal considerations individuals weigh, even with lucrative opportunities.

2. What is the Meta SuperIntelligence Lab and why is it important?

The Meta SuperIntelligence Lab is Meta’s flagship initiative to build artificial general intelligence (AGI). By poaching top talent like Agarwal from rivals, Meta aims to position itself at the cutting edge of AI development, competing directly with Google DeepMind and OpenAI.

3. What lessons can young professionals learn from Agarwal’s career journey?

Agarwal’s story teaches that success in high-tech fields depends on continuous learning, openness to new risks, and aligning work with personal values. High salaries and prestige matter, but so do fulfillment, growth, and a willingness to step out of one’s comfort zone.

References:
Original reporting adapted from LiveMint: Read More

“`
#LLMs #LargeLanguageModels #AI #ArtificialIntelligence #GenerativeAI #MachineLearning #DeepLearning #NLP #AIModels #AITechnology #AITrends #FoundationModels #OpenAI #ChatGPT #AIResearch #AIDevelopment #Transformers #ConversationalAI #AIEthics #AIInnovation