Yash Patil, the 23-year-old behind Applied Compute’s $4.5 billion sale to Palo Alto Networks

At 23, Yash Patil has pulled off one of the fastest exits of the enterprise AI boom. On 18 November 2026, cybersecurity major Palo Alto Networks announced it would acquire Applied Compute, the AI startup he co-founded in May 2025, in an all-stock deal valued at $4.5 billion.

The price came only weeks after Applied Compute was valued at $3.25 billion in a fresh funding round. Once the deal closes, Yash Patil, the company’s co-founder and CEO, will join Palo Alto Networks as Co-Chief AI Officer, bringing the Applied Compute team with him.

Sharing the news on LinkedIn, the Indian-origin founder wrote, “Applied Compute has signed a definitive agreement to be acquired by Palo Alto Networks for $4.5B!”

Yash Patil's post of LinkedIn

From Stanford to OpenAI

Yash studied computer science at Stanford University, where he enrolled in 2021. On campus, he served as co-director of TreeHacks, Stanford’s largest annual hackathon.

In 2023, he joined OpenAI as a Member of Technical Staff in its residency programme, working on post-training infrastructure. Six months later, he became a full-time Member of Technical Staff in San Francisco, working on long-horizon tasks for Codex, OpenAI’s AI-powered coding assistant. He stayed until May 2025.

A bet that every company should own its intelligence

In May 2025, Yash left OpenAI with two fellow researchers to start Applied Compute in San Francisco. Rhythm Garg had been a core contributor to o1, OpenAI’s first reasoning model trained with reinforcement learning, while Linden Li had worked on the systems and infrastructure behind that kind of training. Its software lets businesses take open-weight AI models, refine them after their initial training and run them, all on data the businesses already hold.

Yash Patil with Satya Nadella

The founders’ launch post on the company’s website puts the mission simply, “We’re Applied Compute, and we build Specific Intelligence.” In practice, that means drawing out the know-how already inside a client company, training custom models on it and deploying AI agents that work alongside the client’s staff, belong to the client and keep improving over time. Rather than advising from a distance, Applied Compute’s engineers embed within customer teams.

From zero to $100 million in 18 months

Applied Compute grew at a pace rare even by Silicon Valley standards. It crossed $100 million in annualised revenue run-rate in November 2026, about 18 months after its founding. Its customers include Microsoft, Nvidia, DoorDash, Shopify, Harvey, Cognition and Mercor.

Investors moved just as quickly. A $20 million seed round in June 2025 was followed in April 2026 by $80 million that valued the company at $1.3 billion. This month, a $350 million round led by Elad Gil lifted that figure to $3.25 billion.

Why Palo Alto Networks is paying $4.5 billion

Palo Alto Networks, led by Indian-origin Chairman and CEO Nikesh Arora, expects most large organisations to mix open and closed AI models, and to shape the open ones around their own data and budgets. It plans to use Applied Compute’s technology to help them do that securely, on public clouds or inside their own sovereign infrastructure.

Applied Compute will keep serving its existing customers, while Palo Alto Networks brings the platform to its global customer base. The startup’s researchers will also work on Palo Alto Networks’ in-house AI models, both large and small.

“Applied Compute has built the best post-training and inference platform we’ve seen, and its research team is exceptional,” Nikesh Arora said. “We saw that firsthand as a prospective customer, and our Unit 42 Frontier AI Assessment confirmed their strong security foundations.”

In the official announcement, Yash Patil said, “We built Applied Compute with an exceptional team of AI researchers and engineers to make it easy for any company to train, serve, and improve proprietary models. Joining Palo Alto Networks will put security at the center of all our products and accelerate our traction with the world’s largest organizations.”

The transaction is expected to close in Palo Alto Networks’ fiscal third quarter of 2027, which ends on 30 April 2027, pending regulatory approval.

Applied Compute Co-founders

Applied Compute Co-founders Rhythm, Yash and Linden

Small team, big outcome

For Yash Patil, the milestone belongs to the team as much as to the founders. “Above all, I’m incredibly grateful for the AC team,” he mentioned. “A few dozen people in just over 18 months grew this company from zero revenue to over $100M in annualized revenue run-rate, working with many of the world’s most innovative companies.”

As Co-Chief AI Officer, Yash will now help shape AI strategy at one of the world’s largest cybersecurity companies, barely two years after leaving campus for OpenAI.

ALSO READ: Aman Sanger and Cursor, the MIT-built AI startup now being acquired by Elon Musk’s SpaceX in $60 billion deal

Share with