Method

Feature store

Shared feature definitions with point-in-time and online lookup.

45 cases from 33 companies apply Feature store. 11 were checked in an online A/B test; 21 report results in production.

Typical tasks: Build a data platform, Catch fraud and abuse, Rank search results, Recommend to new users and new items and 13 more. Companies: Pinterest, Airbnb, DoorDash, Netflix, Avito, Whatnot and 27 more. Often combined with: Gradient boosting, Sequential recommendation, Multi-task learning, Rule-based heuristics, Shadow mode and 81 more.

From Data Silos to Feature Serving: How We Unified Fragmented Data for Training, Inference, and Reusesaas-b2b2026-11
Precompute Firstfintech2026-10
Putting a Transformer Inside Every Card Authorizationfintech2026-10
The guest journey, updated in real time: extending Airbnb’s sequence recommender with Chronontravel2026-09
From User Sequences to Scaling Laws: A Multi-Stage Architecture for Meta’s Ads Rankingadtech2026-08
Pinner Progression: Better Use-Case Representation Driving Weekly Active User Growth at Pinterestsocial2026-07
Lakebase and Serving Endpoints Implementation in Real-Time Pricinginsurance2026-06
Ultra Low Latency Streaming for Real-Time Feature Engineeringfintech2026-06
Operating Scalable AI for Games: From LLM Intelligence to Real-Time ML Systemsgaming2026-06
Reliance Jio Customer Churn Radar and next best actiontelecom2026-06
Building a unified consumer memory for personalization at scalefood-delivery2026-06
Thinking Fast & Slow for a Personalized Notification Systemmedia-streaming2026-06
Distilling Long-Tail User Behavior into Scalable Embeddings for Job Searchmarketplace2026-06
ML in Store Operations and Logistics: How Production Models Directly Optimize Processesretail2026-05
AI Assistant for Salesclassifieds2026-05
Enhancing Ad Relevance: Integrating Real-Time Context into Sequential Recommender Modelsadtech2026-05
Integrating Netflix’s Foundation Model into Personalization applicationsmedia-streaming2025-11
The New User Dilemma: Why We Killed Our Heuristics (& What We Built Instead)marketplace2025-09
The evolution of Grab's machine learning feature storefood-delivery2025-07
Contextual Property Embeddings for Corse-grained Personalizationtravel2025-07
ML in AdTech. The Battle for the Click: How We Beat Historical CTR in Advertisingadtech2025-05
Graph Neural Networks in PD for Small and Micro Businessbanking2025-05
Increment vs Experiment: How to Reduce Uncertainty When Developing ML Modelse-commerce2025-05
How Pinterest Accelerates ML Feature Iterations via Effective Backfillsocial2025-05
Evolving Feed Ranking at Whatnotmarketplace2025-04
How to investigate the online vs offline performance for DNN modelsfood-delivery2024-12
Building a User Signals Platform at Airbnbtravel2024-11
Hacker Hunting: Analyzing Process Launch Graphsgeneral2024-11
How We Fought Spam in Reviewsmarketplace2024-05
Feature Script: Speeding Up Development Without Losing Reliabilitybanking2024-05
Shepherd: How Stripe adapted Chronon to scale ML feature developmentfintech2024-04
Real-time Fraud Detection with Yoda and ClickHousemarketplace2024-03
Handling Online-Offline Discrepancy in Pinterest Ads Ranking Systemadtech2024-01
From Pre-trained to Fine-tuned: Nextdoor’s Path to Effective Embedding Applicationssocial2023-09
How Instacart Modernized the Prediction of Real Time Availability for Hundreds of Millions of Items While Saving Costse-commerce2023-07
How we built our Machine Learning Pipeline to fight fraud at BlaBlaCar — Part 2mobility2023-06
Real-Time Personalization Using Microservicesretail2023-06
ML System Design Doc Challenge – Finalbanking2023-05
The Journey to Machine-Learned Re-rankingmarketplace2023-01
Evolution of Search Ranking at Thumbtackmarketplace2022-12
Search Pipeline: Part IIgeneral2022-11
Speed Up Feature Engineering for Recommendation Systemssocial2022-09
Graph Machine Learning at Airbnbmarketplace2022-06
Modernizing Nextdoor Search Stack — Part 2social2022-05
Near real-time features for near real-time personalizationsocial2022-03