What Is eugeneyan.com and Who Runs It?

Updated

eugeneyan.com is the personal site of Eugene Yan, a Member of Technical Staff at Anthropic. It collects his writing, talks, and prototype projects on recommendation systems, LLMs, and AI-powered products, and serves as the entry point to a newsletter he sends to more than 11,800 readers. It is a personal publication and portfolio, not a company product site.

Who runs the site

Eugene Yan describes his work as bridging "the field and the frontier" and helping build safe, reliable AI systems that scale. According to the site, he is currently a Member of Technical Staff at Anthropic, working on recommendation systems and AI-powered products, and he has previously led ML/AI teams at Amazon, Alibaba, Lazada, and a Healthtech Series A company.

What you'll find there

The site is organized into a few main areas:

  • Writing — long-form posts on RecSys, LLMs, evaluation, and engineering practice.
  • Speaking — conference talks and keynotes.
  • Prototyping — small apps he builds for fun, such as AI Reading Club, AlignEval, Obsidian Copilot, and AI Coach.
  • About / Start Here — background and suggested entry points.

He also teaches applied ML and LLM material through two companion sites, ApplyingML.com and Applied-LLMs.org, both linked from the homepage.

Representative articles

The homepage lists recent and popular posts, which give a good sense of the site's range:

Date Title
21 Jun 2026 Patterns for Building Cybersecurity Evals
27 May 2026 Using LLMs to Secure Source Code
03 May 2026 How to Work and Compound with AI
23 Nov 2025 Product Evals in Three Simple Steps
14 Sep 2025 Training an LLM-RecSys Hybrid for Steerable Recs with Semantic IDs
18 Aug 2024 Evaluating the Effectiveness of LLM-Evaluators (aka LLM-as-Judge)
07 Jul 2024 How to Interview and Hire ML/AI Engineers
30 Jul 2023 Patterns for Building LLM-based Systems & Products

Several of these are marked as having more than 50k reads, including the LLM-evaluator and LLM-system-patterns pieces.

Talks and newsletter

Featured talks include "Improving RecSys & Search with LLM techniques" at AI Engineer 2025, a keynote at AI Engineer 2024, and a session at Netflix PRS 2024. The newsletter covers RecSys, LLMs/AI, and what he is learning; the site provides a signup link for email updates.

When this site is useful to you

If you work on recommendation systems or LLM-based products and want practical, experience-based write-ups on evaluation, system patterns, and hiring, the site is a reasonable place to start — begin with the "Start Here" page or the popular-posts list. If you are looking for a product, pricing, or support channel, this is not that kind of site; it is one practitioner's public notebook and teaching outlet.