Hi, I'm Eugene Yan,
I build recommendation systems and AI-powered experiences that serve customers at scale. Currently, I'm a Principal Applied Scientist at Amazon. Outside of work, I also...
- Write and speak about recommendation systems, AI-powered product, and engineering.
- Prototype apps like AI Reading Club, AlignEval, Obsidian Copilot, AI Coach, and more.
- Teach how to apply ML/LLMs effectively at ApplyingML.com & Applied-LLMs.org.
- Send a newsletter on RecSys, LLMs/AI, and what I'm learning to 11,800+ readers.
19 Oct 2025· Advice for New Principal Tech ICs (i.e., Notes to Myself)14 Sep 2025· Training an LLM-RecSys Hybrid for Steerable Recs with Semantic IDs22 Jun 2025· Evaluating Long-Context Question & Answer Systems04 Jun 2025· AI Engineer 2025 - Improving RecSys & Search with LLM techniques18 May 2025· Exceptional Leadership: Some Qualities, Behaviors, and Styles04 May 2025· Building News Agents for Daily News Recaps with MCP, Q, and tmux20 Apr 2025· An LLM-as-Judge Won't Save The Product—Fixing Your Process Will30 Mar 2025· Frequently Asked Questions about My Writing Process18 Mar 2025· NVIDIA GTC 2025 - Building LLM-Powered Applications16 Mar 2025· Improving Recommendation Systems & Search in the Age of LLMs
Trending This Week
18 Aug 2024· Evaluating the Effectiveness of LLM-Evaluators (aka LLM-as-Judge)26 May 2024· Prompting Fundamentals and How to Apply them Effectively12 May 2024· What We've Learned From A Year of Building with LLMs30 Jul 2023· Patterns for Building LLM-based Systems & Products11 Jun 2023· Obsidian-Copilot: An Assistant for Writing & Reflecting21 May 2023· Some Intuition on Attention and the Transformer14 Aug 2022· Simplicity is An Advantage but Sadly Complexity Sells Better31 Jul 2022· Uncommon Uses of Python in Commonly Used Libraries12 Jun 2022· Design Patterns in Machine Learning Code and Systems
18 Mar 2025· NVIDIA GTC 2025 - Building LLM-Powered Applications27 Jun 2024· AI Engineer 2024 Keynote - What We Learned from a Year of LLMs31 May 2024· Netflix PRS 2024 - Applying LLMs to Recommendation Experiences09 Oct 2023· AI Engineer 2023 Keynote - Building Blocks for LLM Systems23 Sep 2022· RecSys 2022 Keynote - Is the Juice Worth the Squeeze?
12 Jan 2025· Building AI Reading Club: Features & Behind the Scenes27 Oct 2024· AlignEval: Building an App to Make Evals Easy, Fun, and Automated04 Sep 2022· Writing Robust Tests for Data & Machine Learning Pipelines
208 posts, 31 talks, 19 prototypes, 419,608 words, and countless hours.
• News Agents: Automating my daily news flash via agentic workflows
• AI Reading Club: Prototyping an AI-powered reading experience
• AlignEval: An App to Make Evals Easy, Fun, and Semi-Automated
• AI Coach: Talk to Tara, an AI Coach, anytime at +1 (206) 558 8782
• Obsidian-Copilot: A Prototype Assistant for Writing and Reflecting
• Raspberry-LLM: Dr. Seuss headlines, HackerNews trolls, etc.
• ApplyingML.com: Ghost knowledge of applying ML effectively
• RecSys in PyTorch: Baselines + improvements via Graphs & NLP
• Image Classification: Transfer learning via Keras and Theano
• Title Classification: Getting Amazon data, modeling, building a UI
Resources
- •
applied-llms.org: Practical lessons from a year of building with LLMs - •
applyingml.com: Papers, guides, and interviews on how to apply ML effectively - •
applied-ml: Papers on real-world machine learning systems in industry - •
open-llms: Open large language models available for commercial use - •
llm-reading-list: Language modeling reading list (to start a paper club) - •
ml-design-docs: Template of design docs for machine learning systems - •
- •
python-collab-template: Template with tests, type checks, linting, etc. - •
papermill-mlflow: Experimentation workflow for machine learning - •
1-on-1s: Questions to ask during 1-on-1s, from my time as a manager - •
omscs-faq: Georgia Tech Online Master's of Science in Computer Science - •
teardowns: Surveys & deep dives of data/machine learning systems