# Tag: recsys

> Posts and talks on recommender systems.

- URL: https://eugeneyan.com/tag/recsys
- Index of the whole site: https://eugeneyan.com/llms.txt

- [Training an LLM-RecSys Hybrid for Steerable Recs with Semantic IDs](https://eugeneyan.com/writing/semantic-ids.md) (2025-09-14): An LLM that can converse in English & item IDs, and make recommendations w/o retrieval or tools.
- [AI Engineer 2025 - Improving RecSys & Search with LLM techniques](https://eugeneyan.com/speaking/aie-2025.md) (2025-06-04): Recsys & search are converging with LLMs via semantic IDs, data augmentation, and unified foundation models.
- [Improving Recommendation Systems & Search in the Age of LLMs](https://eugeneyan.com/writing/recsys-llm.md) (2025-03-16): Model architectures, data generation, training paradigms, and unified frameworks inspired by LLMs.
- [Push Notifications: What to Push, What Not to Push, and How Often](https://eugeneyan.com/writing/push.md) (2023-12-24): Sending helpful & engaging pushes, filtering annoying pushes, and finding the frequency sweet spot.
- [More Design Patterns For Machine Learning Systems](https://eugeneyan.com/writing/more-patterns.md) (2023-04-23): 9 patterns including HITL, hard mining, reframing, cascade, data flywheel, business rules layer, and more.
- [RecSys 2022: Recap, Favorite Papers, and Lessons](https://eugeneyan.com/writing/recsys2022.md) (2022-10-02): My three favorite papers, 17 paper summaries, and ML and non-ML lessons.
- [RecSys 2022 Keynote - Is the Juice Worth the Squeeze?](https://eugeneyan.com/speaking/recsys2022-keynote.md) (2022-09-23): Invited keynote at the Workshop for Online Recommender Systems and User Modeling (ORSUM)
- [Bandits for Recommender Systems](https://eugeneyan.com/writing/bandits.md) (2022-05-08): Industry examples, exploration strategies, warm-starting, off-policy evaluation, and more.
- [How to Measure and Mitigate Position Bias](https://eugeneyan.com/writing/position-bias.md) (2022-04-17): Introducing randomness and/or learning from inherent randomness to mitigate position bias.
- [Counterfactual Evaluation for Recommendation Systems](https://eugeneyan.com/writing/counterfactual-evaluation.md) (2022-04-10): Thinking about recsys as interventional vs. observational, and inverse propensity scoring.
- [RecSys 2021 - Papers and Talks to Chew on](https://eugeneyan.com/writing/recsys2021.md) (2021-10-03): Simple baselines, ideas, tech stacks, and packages to try.
- [MLOps Community - System Design for RecSys & Search](https://eugeneyan.com/speaking/mlops-community-recsys.md) (2021-09-15): An overview of system design, candidate retrieval, and ranking, with industry examples.
- [Reinforcement Learning for Recommendations and Search](https://eugeneyan.com/writing/reinforcement-learning-for-recsys-and-search.md) (2021-09-05): Focusing on long-term rewards, exploration, and frequently updated item.
- [SF Big Analytics - System Design for RecSys & Search](https://eugeneyan.com/speaking/sf-data-meetup-recsys.md) (2021-07-13): Why real-time RecSys? What does the system design look like in industry? How to build an MVP?
- [System Design for Recommendations and Search](https://eugeneyan.com/writing/system-design-for-discovery.md) (2021-06-27): Breaking it into offline vs. online environments, and candidate retrieval vs. ranking steps.
- [Patterns for Personalization in Recommendations and Search](https://eugeneyan.com/writing/patterns-for-personalization.md) (2021-06-13): A whirlwind tour of bandits, embedding+MLP, sequences, graph, and user embeddings.
- [Real-time Machine Learning For Recommendations](https://eugeneyan.com/writing/real-time-recommendations.md) (2021-01-10): Why real-time? How have China & US companies built them? How to design & build an MVP?
- [RecSys 2020: Takeaways and Notable Papers](https://eugeneyan.com/writing/recsys2020.md) (2020-09-27): Emphasis on bias, more sequential models & bandits, robust offline evaluation, and recsys in the wild.
- [Serendipity: Accuracy’s Unpopular Best Friend in Recommenders](https://eugeneyan.com/writing/serendipity-and-accuracy-in-recommender-systems.md) (2020-04-26): What I learned about measuring diversity, novelty, surprise, and serendipity from 10+ papers.
- [DataScience SG Meetup - RecSys, Beyond the Baseline](https://eugeneyan.com/speaking/recommender-systems-beyond-the-baseline-talk.md) (2020-01-14): Comparing baselines (matrix factorization) against novel approaches using graphs & NLP.
- [Beating the Baseline Recommender with Graph & NLP in Pytorch](https://eugeneyan.com/writing/recommender-systems-graph-and-nlp-pytorch.md) (2020-01-13): Beating the baseline using Graph & NLP techniques on PyTorch, AUC improvement of ~21% (Part 2 of 2).
- [Building a Strong Baseline Recommender in PyTorch, on a Laptop](https://eugeneyan.com/writing/recommender-systems-baseline-pytorch.md) (2020-01-06): Building a baseline recsys based on data scraped off Amazon. Warning - Lots of charts! (Part 1 of 2).
