Method

Two-stage recommender

Candidate generation followed by a ranking model.

28 cases from 18 companies apply Two-stage recommender. 12 were checked in an online A/B test; 5 report results in production.

Typical tasks: Retrieve and rank from millions of items, Recommend to new users and new items, Balance relevance and diversity, Rank search results and 5 more. Companies: Pinterest, Meta, DoorDash, X5 Tech, Criteo, Whatnot and 12 more. Often combined with: A/B test, Learning to rank, Result diversification, Two-tower model, Gradient boosting and 50 more.

Your Algorithm: Giving People More Ways to Personalize Facebook Recommendationssocial2026-11
Meta Andromeda: Powering the Next Generation of Ads Retrieval Through Model-System Co-Designadtech2026-10
Turning Opportunistic AI Inference Capacity Into Better Ads Recommendationsadtech2026-10
Beyond Two Towers: Launching the 3-Tower Engagement Co-Train Model (Part 2)adtech2026-09
A Reranker in Retail: Turning Advertising into Recommendationsretail2026-05
Evolution of Multi-Objective Optimization at Pinterest Home feedsocial2026-04
Adapting the Facebook Reels RecSys AI Model Based on User Feedbacksocial2026-01
How RecSys & LLMs Will Converge: Architecture of Hybrid RecoAgentsadtech2025-10
Using LLMs to infer grocery preferences from DoorDash restaurant ordersfood-delivery2025-09
Modernizing Home Feed Pre-Ranking Stagesocial2025-05
Establishing a Large Scale Learned Retrieval System at Pinterestsocial2025-01
Meta Andromeda: Supercharging Advantage+ automation with the next-gen personalized ads retrieval engineadtech2024-12
Whatamix: Blendable feed constructionmarketplace2024-11
Channel-Smart Property Search: How Expedia Tailors Rankings for Youtravel2024-07
Building a Large-Scale Recommendation System: People You May Knowsocial2024-02
Improving the OpenChat recommendation model through offline and online A/B testingsocial2024-01
Personalizing the DoorDash Retail Store Page Experiencefood-delivery2023-12
Conference Recap: ECIR23 Take-awaysmedia-streaming2023-08
How We Learned to Guess Shoppers' Tastes and Made Money on It: Personalized Catalog Ranking at Lamodae-commerce2023-05
Representation online matters: practical end-to-end diversification in search and recommender systemssocial2023-05
Query Rewards: Building a Recommendation Feedback Loop During Query Selectionsocial2022-10
Recommend API: one shared pipeline for many recommendation features at Slacksaas-b2b2022-09
How we sped up shipping new features in rankingfood-delivery2022-06
MegaFon RecSys Workshop, part 2: a two-level recommender (LightFM + CatBoost) and DSSM modelsmedia-streaming2022-06
Modernizing Nextdoor Search Stack — Part 2social2022-05
Beyond Matrix Factorization: Using hybrid features for user-business recommendationsmarketplace2022-04
Improving Shopping Recommendations for Customers Through eBay’s Relevance Cascade Modele-commerce2022-04
Building a Platform for Serving Recommendations at Etsymarketplace2022-03