Eugene Yan is a Principal Applied Scientist at Amazon, building AI-powered recommendation systems and experiences. He shares insights on RecSys, LLMs, and applied machine learning, while mentoring and investing in ML startups.
185 articles from this blog
Explores counterfactual evaluation as an alternative to A/B testing for offline assessment of recommendation systems.
Explores the difference between high-level intent and low-level requirements in software engineering and how to effectively navigate between them.
A data scientist shares a structured approach to starting data science projects, focusing on business goals, requirements, and avoiding common pitfalls.
A data leader shares advice on creating a vision and roadmap for a data team, including stakeholder engagement and problem evaluation.
Key warning signs for data scientists and tech professionals to consider before joining a new data team, covering data infrastructure, team roadmap, and role clarity.
Practical strategies for staying current in the fast-moving field of machine learning, including project experimentation and community engagement.
A podcast interview discussing common ML mistakes, quantifying impact, and career growth for machine learning engineers.
The author introduces ApplyingML.com, a site dedicated to sharing practical knowledge and interviews on applying machine learning effectively in real-world work.
A guest author shares five key lessons learned from writing a technical blog, focusing on overcoming perfectionism and getting started.
A summary of key papers and talks from the RecSys 2021 conference, focusing on collaborative filtering, model comparisons, and deployment strategies.
Advises starting ML projects with simple heuristics and data analysis before implementing complex machine learning models, citing expert advice.
A talk on system design principles for building production recommendation systems and search engines, presented at an MLOps Community meetup.
Explores how reinforcement learning methods like bandits and policy-based approaches can improve recommendation systems by optimizing for long-term rewards.
Profile of Amazon applied scientist Eugene Yan, focusing on his career in data science and his influential technical writing about machine learning.
Explores methods like semi-supervised and active learning to create training labels when labeled datasets are unavailable, with industry examples.
Explains how to bootstrap training labels for a semantic search system using initial lexical search and user click data instead of costly human annotation.
A talk on system design for recommendation and search systems, covering architecture and production considerations.
A data scientist shares practical strategies and mindsets for influencing technical teams and driving change without formal authority.
Explores system design patterns for industrial-scale recommendation and search engines, focusing on offline/online components and retrieval/ranking stages.
Explores machine learning patterns like bandits, sequential, and graph-based models for personalizing recommendations and search results.
