“`html

IIT Bombay Alumnus Leaves ₹8 Crore Meta Job for Startup Dream

TL;DR

Rishabh Agarwal, an IIT Bombay and Mila AI PhD graduate, shocked the tech world by quitting his ₹8 Crore (over $1M) annual salary job at Meta’s elite AI ‘Superintelligence’ lab after just 5 months. He’s leaving to chase his own entrepreneurial ambitions. This move highlights the growing trend of top AI talent leaving lucrative Big Tech positions for riskier, but more fulfilling, startup ventures.

The Tech World Reacts: Why Would Anyone Leave a ₹8 Crore Meta AI Job?

In a landscape where top AI engineers and researchers are being chased by tech giants with lucrative pay packages, it’s rare to see an individual voluntarily walk away – especially after being part of groundbreaking projects at places like Google Brain, DeepMind, and Meta’s Superintelligence team. And yet, Rishabh Agarwal did just that in August 2025.

Only five months after joining Meta’s core AI group—the same department Meta CEO Mark Zuckerberg has publicly called the company’s “top priority”—Rishabh chose passion and uncertainty over security and prestige.

Who is Rishabh Agarwal?

• Alumnus of IIT Bombay—secured AIR 33 in JEE
• PhD in Computer Science at Mila – Quebec AI Institute—affiliated with renowned AI expert Yoshua Bengio
• Internship stints at Tower Research Capital, Saavn, Waymo
• Industry Experience: Senior Research Scientist at Google Brain, Staff Research Scientist at Google DeepMind, Research Scientist at Meta Superintelligence Lab
• Joined Meta’s newly formed Superintelligence division in April 2025

Salary: News reports peg his CTC at around ₹8 Crore per annum, which is $1M+ (including stock options, bonuses, etc.) – one of the highest salary bands for technical employees in the world.

What Prompted Rishabh’s Decision?

Rishabh made the announcement directly in his own words via X (formerly Twitter):

“This is my last week at @AIatMeta. It was a tough decision not to continue with the new Superintelligence 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.”

He credits the vision and recruiting pitch from Mark Zuckerberg (Meta CEO) and Alexander Wang (Chief AI Officer) as “compelling”, but ultimately, he’s following Mark’s own advice that “in a world that’s changing so fast, the biggest risk you can take is not taking any risk.”

• Desire to build and create: Rishabh alludes to wanting more “direct risk exposure” and experience as a creator or founder.
• Longing for new challenges: After almost 8 years at the world’s top research labs, he seeks “a different kind of risk” and personal growth.
• Inspired by mentors and industry trends: As AI rapidly evolves and paradigm shifts beckon (like agentic AI, open-source LLMs, and new startup opportunities), many in the field are feeling a pull toward entrepreneurship.

Inside Meta’s Superintelligence Lab: What Rishabh Achieved

Meta’s AI division, especially its superintelligence unit, is at the forefront of cutting-edge research in areas like reinforced learning (RL), foundation models, and synthetic data pipelines. Rishabh, in his departure note, summarized his major contributions:

• Pushed 8-bit dense models to approach DeepSeek R1 performance (a notable rival in the field) with scalable reinforcement learning techniques.
• Developed synthetic data mid-training methods to warm-start reinforcement learning models for faster, more robust training cycles.
• Innovated improved on-policy distillation methods—crucial for compressing large models into smaller, practical versions for deployment.

Meta’s AI and machine learning division has been vying for global leadership, reportedly hiring experts with multi-million dollar annual packages. Competing against OpenAI, Google, and DeepMind, Meta’s effort to build “superintelligent” AI platforms is a central focus for Zuckerberg and has attracted some of the world’s brightest minds.

The Allure—and Pressure—of Big Tech AI Jobs

In 2025, “AI superstars” are as sought-after as top athletes or Hollywood actors. Here’s why many would have seen Rishabh’s Meta role as a “dream job”:

• Unmatched resources: Access to unlimited computing power (Nvidia H100 clusters, dedicated research budgets, large datasets).
• Collaboration with world-class peers – every team member is a pioneer in AI, ML, NLP, or RL.
• Influence: Work that impacts billions—Meta’s AI algorithms power Facebook, Instagram, WhatsApp, and more.
• Security and status: Lucrative salary, stock options, health insurance, visa sponsorships – financial freedom and a global reputation.

But there’s another side:

• Bureaucracy and organizational fatigue—even high-impact researchers are often several layers removed from real product decisions.
• Pace: Large, public companies move slower than startups—research that “moves the needle” might take years to reach real-world users.
• Personal passion: After years accumulating skills and resources, some feel an urge to chart their own course and shape the next wave of AI innovation directly—rather than as a cog within huge organizations.

A Broader Trend: The “AI Brain Drain” Toward Startups

Rishabh Agarwal’s story is not unique. In the last three years, there has been surge of high-profile “Big Tech” AI experts walking away from top-tier jobs to launch their own ventures or join early-stage startups:

• Ilya Sutskever (OpenAI cofounder, formerly Google Brain) left to shape the fore of generative AI.
• Boris Power & others exited DeepMind, Anthropic, OpenAI for entrepreneurial journeys.
• Grok AI, Mistral, Inflection, and more – all founded by alumni of Google, Meta, or OpenAI, now raising hundreds of millions in funding to shape the future of artificial intelligence.

Why is this “AI brain drain” accelerating?

• AI startup opportunities are expanding rapidly: From agentic AI (tools that act autonomously), developer platforms, synthetic data, finetuning, “AI as a Service”, to vertical-specific LLMs for healthcare, law, and education.
• Funding is abundant: 2024–2025 has seen record capital deployment from VCs, sovereign funds, and even governments eager to bet big on disruptive AI startups.
• Cultural moment: The public hype and adoption of AI (thanks to ChatGPT, Gemini, xAI, etc.) has created a gold-rush mindset.
• Personal calling: For many, the ability to “bet on yourself” and try to create lasting impact outweighs even a multi-crore pay package.

Rishabh’s Inspiration: Mark Zuckerberg’s Philosophy

In his farewell note, Rishabh highlights how Mark Zuckerberg’s guidance influenced his decision:

“In a world that’s changing so fast, the biggest risk you can take is not taking any risk.”

This resonates with many in the tech world. With AI, the rules change every few months. Those who stagnate risk being left behind, while those who leap—even from a place of comfort—are the ones who push the world forward.

What’s Next for Rishabh Agarwal?

While he hasn’t officially announced his new venture yet, Rishabh’s background sets him up for success in nearly any AI startup domain:

• Core ML/AI infrastructure tools
• Enterprise AI platforms
• Specialized synthetic data pipelines
• Consumer LLM-based apps

What to expect: Given his academic and industry pedigree, he’ll likely attract capital and top talent rapidly. His journey will be watched closely by aspiring engineers, investors, and tech enthusiasts across India and internationally.

Key Lessons: Flipping the Narrative on Risk

• Security isn’t always the best bet—sometimes, walking away from comfort creates bigger opportunities to learn, build, and lead.
• In modern AI, fast cycles and autonomy are key—for those with the right expertise, the startup path enables direct ownership and impact.
• India’s tech talent is on the global map—stories like Rishabh’s will inspire others from IITs and premier institutes to not just take big jobs, but to chase big dreams.

Conclusion: A Symbol for India’s Next-Gen Innovators

Rishabh Agarwal’s journey is emblematic of a generation of Indian engineers who are no longer just grabbing spots at the world’s biggest companies—they’re building the companies, products, and even the next generation of AI breakthroughs themselves.

As he embarks on his entrepreneurial journey, his high-profile exit will ignite both curiosity and courage among India’s aspiring founders. It isn’t just about the money—it’s about meaning, impact, and personal growth.

For those watching: Keep an eye on Rishabh’s next move. In today’s fast-changing AI world, sometimes the “biggest risk” is the one that leads to the next billion-dollar idea.

FAQs About Rishabh Agarwal’s Meta Exit

1. Why did Rishabh Agarwal leave his crorepati Meta job?

Answer: He wanted to pursue a “different kind of risk”—building something on his own, likely a startup—after years at Big Tech giants. He felt that staying would mean missing the opportunity at a high-potential moment for AI entrepreneurs.

2. What did Rishabh do at Meta’s Superintelligence lab?

Answer: Rishabh contributed to developing scalable 8-bit AI models, innovated with synthetic data for RL training, and improved distillation techniques—helping Meta stay competitive in the global AI research race.

3. What is the broader trend from this story?

Answer: There’s a growing wave of elite AI researchers leaving lucrative jobs at tech giants to start their own companies, ride the AI gold rush, and shape the direction of the next generation of artificial intelligence.

For more stories about India’s tech trailblazers and the future of artificial intelligence, keep following our blog!

“`
#LLMs #LargeLanguageModels #AI #ArtificialIntelligence #GenerativeAI #MachineLearning #DeepLearning #NLP #AIModels #AIEthics #AIFuture #AITrends #FoundationModels #AIEvolution #AIResearch