How Pinterest Improved Analytics Agent Compliance with Loop Engineering |  | social | 2026-11 |
From Data Silos to Feature Serving: How We Unified Fragmented Data for Training, Inference, and Reuse |  | saas-b2b | 2026-11 |
Your Algorithm: Giving People More Ways to Personalize Facebook Recommendations |  | social | 2026-11 |
Fact-Checking at Scale: Building a More Accurate and Efficient Synopsis Factuality System |  | media-streaming | 2026-11 |
Safe and Efficient Multi-Epoch Embedding Training |  | social | 2026-11 |
Composing Fraud Rules at the Speed of the Attack |  | fintech | 2026-11 |
Meta Andromeda: Powering the Next Generation of Ads Retrieval Through Model-System Co-Design |  | adtech | 2026-10 |
How much should we delegate to an LLM? Automating advertising analytics reports with an AI agent |  | adtech | 2026-10 |
Agentic Credit Analysis |  | fintech | 2026-10 |
PredictPay: From Hackathon Idea to Smarter Payments at Scale |  | fintech | 2026-10 |
GenRecs Tokenization: Transforming Ads Recommendation at Meta with Stable Model Vocabulary |  | adtech | 2026-10 |
From Activity to Intent: Generating User Journeys with LLMs |  | social | 2026-10 |
Measuring the Impact of Personalized Recommendations |  | media-streaming | 2026-10 |
DataJunction as Netflix’s answer to the missing piece of the modern data stack |  | media-streaming | 2026-10 |
Precompute First |  | fintech | 2026-10 |
Meta’s Generative Ads Model (GEM): Scaling the Central Model and Compute Engine for Frontier Ads AI |  | adtech | 2026-10 |
CASTLE: training natural language search before the first real query |  | travel | 2026-10 |
Turning Opportunistic AI Inference Capacity Into Better Ads Recommendations |  | adtech | 2026-10 |
Inside LinkedIn's data context layer: curating trustworthy context for humans and agents |  | social | 2026-10 |
Workbench: Building an AI Content Generation and Quality Assurance System at Scale |  | marketplace | 2026-10 |
Putting a Transformer Inside Every Card Authorization |  | fintech | 2026-10 |
Metrics Board: Building an Agent-ready Metrics Layer |  | social | 2026-10 |
Building Gavel: AI evaluation infrastructure across Eternal |  | food-delivery | 2026-10 |
Powering AI-led research through simulation |  | food-delivery | 2026-10 |
Unsupervised Topic Discovery with Text Embeddings |  | fintech | 2026-10 |
Advancing Generative Recommendation Inference: our collaboration with NVIDIA on TensorRT LLM |  | media-streaming | 2026-10 |
Personalization without user identity |  | travel | 2026-09 |
The Model Is Not the Modeling |  | fintech | 2026-09 |
Automated merchant review summary system with integrated feedback |  | food-delivery | 2026-09 |
Roonomics: How Economics Helped Us to Build a Churn Prediction Model |  | food-delivery | 2026-09 |
The guest journey, updated in real time: extending Airbnb’s sequence recommender with Chronon |  | travel | 2026-09 |
Beyond Two Towers: Launching the 3-Tower Engagement Co-Train Model (Part 2) |  | adtech | 2026-09 |
Learning Richer Credit Representations with Transformers at Affirm |  | fintech | 2026-09 |
Inside Vera, DoorDash’s Data Agent |  | food-delivery | 2026-09 |
ML based ranking using Nrtsearch |  | search-engine | 2026-09 |
Building a Transformer-Based Category Recommender at Thumbtack |  | marketplace | 2026-09 |
Refreshing the Travel-Time Map Behind Lyft’s Marketplace: Rebuilding Neighborhood Reachability Signals |  | mobility | 2026-09 |
The right fit: building a parts intelligence solution for Shopify |  | e-commerce | 2026-09 |
The Lifecycle of LLM-as-a-Judge: Building, Aligning, and Monitoring at scale |  | media-streaming | 2026-09 |
How to improve Algolia search quality for content-heavy sites |  | general | 2026-09 |
Why Spotify Is Not Using Bayesian A/B Testing |  | media-streaming | 2026-09 |
MAPS: Netflix’s Multimodal Asset Personalization at Scale |  | media-streaming | 2026-08 |
Eval-driven development: Lessons from evaluating GenAI at scale |  | travel | 2026-08 |
Developing a model to detect harmful content from OpenChat titles and descriptions |  | social | 2026-08 |
Rebuilding LinkedIn's Follows Recommendations with LLM-Based Semantic Retrieval and Ranking |  | social | 2026-08 |
Building Jarvis Pro: Route first, answer later |  | food-delivery | 2026-08 |
Automating Search Relevance Assessment at Scale with LLM-as-a-Judge |  | e-commerce | 2026-08 |
How we knew COVID was over (and what our models had to unlearn) |  | travel | 2026-08 |
| Calibrating LLM-Based Population Estimates with Human Validation | | marketplace | 2026-08 |
How We’re Building Scam Alert on WhatsApp With End-to-End Encryption and Verifiability Guarantees |  | social | 2026-08 |