# Sahil Agarwal

> Sahil Agarwal is the co-founder and CEO of Enkrypt AI, a Boston-based startup building a security and compliance control layer for enterprise generative AI. An applied mathematician with a PhD from Yale who once modeled the chaotic motion of Arctic sea ice, he now leads a team that red-teams large language models to find where they break - work that made headlines when Enkrypt published research showing DeepSeek-R1 was 11x more likely to generate harmful content than its rivals. He met co-founder Prashanth Harshangi during their PhD years at Yale, and together they raised a $2.35M seed round to make AI safe enough for Fortune 500 deployment.

- **Role:** Co-Founder & CEO at Enkrypt AI
- **Organizations:** Enkrypt AI, Accrete AI, Yale University, University of Oxford
- **Nationality:** Indian
- **Education:** BTech, Mathematics and Computing, Indian Institute of Technology Guwahati, PhD, Applied Mathematics, Yale University, Postgraduate / doctoral research, University of Oxford
- **Known for:** Co-founded and leads Enkrypt AI, an enterprise AI security and compliance platform, Raised $2.35M seed funding led by Boldcap with Berkeley SkyDeck, Kubera VC, Arka VC and others, Authored 10+ publications in international scientific journals and holds 3 patents

## Career timeline

- **—** — Earned BTech in Mathematics and Computing from IIT Guwahati
- **—** — Completed PhD in Applied Mathematics at Yale University, where he met co-founder Prashanth Harshangi
- **—** — Conducted research at University of Oxford on nonlinear dynamics and stochastic processes
- **—** — Led the AI team at Accrete AI as Chief AI Officer
- **2022** — Co-founded Enkrypt AI with Prashanth Harshangi
- **2024** — Enkrypt AI raised $2.35M seed round led by Boldcap
- **2025** — Published red-team research showing DeepSeek-R1 was 11x more likely to generate harmful content

## Achievements

- Co-founded and leads Enkrypt AI, an enterprise AI security and compliance platform
- Raised $2.35M seed funding led by Boldcap with Berkeley SkyDeck, Kubera VC, Arka VC and others
- Authored 10+ publications in international scientific journals and holds 3 patents
- Research recognized by the American Physical Society, Yale Scientific Magazine, and Yale News
- Published widely-cited red-team study on DeepSeek-R1's safety vulnerabilities
- Trusted advisor to Fortune 500 companies on safe AI deployment

## Latest updates

- **2025-01** — Enkrypt AI published red-team research finding DeepSeek-R1 11x more likely to generate harmful content, drawing global coverage.
- **2024-02** — Closed a $2.35M seed round to build a visibility and security layer for generative AI.

## Links

- Website: https://enkryptai.com
- LinkedIn: http://www.linkedin.com/in/sahil-agarwal

---

Profile page: https://yespress.io/sahil-agarwal
Published by YesPress — https://yespress.io
Last updated: 2026-06-21