简体中文
EnglishWhat Is Eugene Yan's Professional Background and Speaking Experience?
Updated
Eugene Yan is currently a Member of Technical Staff at Anthropic, where he builds recommendation systems and AI-powered products at scale. Before that, he led ML/AI teams at Amazon, Alibaba, Lazada, and a Healthtech Series A company. His public speaking centers on applying LLMs to recommendation and search, with appearances at AI Engineer 2024 and 2025 and Netflix PRS 2024.
Current Role and Professional Background
According to his site, Yan describes his work as bridging "the field and the frontier" to help build safe, reliable AI systems that scale. His current position at Anthropic involves recommendation systems and AI-powered products serving customers at scale.
His prior experience spans several organizations where he led ML/AI teams:
| Organization | Role |
|---|---|
| Anthropic | Member of Technical Staff (current) |
| Amazon | Led ML/AI teams |
| Alibaba | Led ML/AI teams |
| Lazada | Led ML/AI teams |
| Healthtech Series A company | Led ML/AI teams |
Beyond his day job, he writes and speaks about AI-powered products, recommendation systems, and engineering. He also teaches how to apply ML/LLMs effectively through ApplyingML.com and Applied-LLMs.org, and runs a newsletter on RecSys, LLMs/AI, and his ongoing learning that reaches 11,800+ readers.
Speaking Experience
Yan's featured talks focus on the intersection of LLMs and recommendation systems:
- AI Engineer 2025 (June 4, 2025) — "Improving RecSys & Search with LLM techniques"
- AI Engineer 2024 Keynote (June 27, 2024) — "What We Learned from a Year of LLMs"
- Netflix PRS 2024 (May 31, 2024) — "Applying LLM..."
The recurring theme across these talks is how LLM techniques can improve recommendation systems and search — a topic he also covers extensively in his writing, including his widely read piece "Improving Recommendation Systems & Search in the Age of LLMs" (March 16, 2025).
What He Writes About
His published work on eugeneyan.com covers several connected areas:
- LLM systems and products — patterns for building LLM-based systems, prompting fundamentals, and lessons from a year of building with LLMs
- Evaluation — LLM-as-judge effectiveness, product evals, long-context Q&A evaluation, and cybersecurity evals
- Recommendation systems — LLM-RecSys hybrids, semantic IDs, and search improvements
- Engineering and career — interviewing and hiring ML/AI engineers, advice for principal tech ICs, and leadership qualities
He also builds prototype applications such as AI Reading Club, AlignEval, Obsidian Copilot, and AI Coach.
Where to Follow His Work
- Writing and talks: eugeneyan.com
- Newsletter: email updates on RecSys, LLMs/AI, and what he's learning (11,800+ readers)
- Teaching: ApplyingML.com and Applied-LLMs.org
If you want to understand how LLMs are being applied to recommendation and search in production settings, his AI Engineer talks and the corresponding write-ups are the most direct entry points. For broader LLM system-building patterns, his 2023 piece "Patterns for Building LLM-based Systems & Products" remains one of his most-read articles.