Case studies / Ads

• Meta11 Nov 2026NewAds

  Intelligent Delivery: Making Every Unit of Compute Serve Better Ads

  Problem Ads ranking needed more compute than available, hurting performance and latency.

  Solution Created an ML‑driven adaptive layer that predicts each request’s value and routes compute to the most useful models, hardware, and stages in real time.

  ≈10% ranking performance boost

  • Cost
  • Latency
  • GPUs

  Read on engineering.fb.comDetailsSave
• Meta3 Nov 2026Ads

  F3 Lightning: How Meta Accelerates Ads ML Features to Production at Scale

  Problem Feature extraction duplicated work, causing slow and costly ad model predictions.

  Solution Created a DAG compiler that merges feature graphs, sharing common operations so each computation runs only once per request.

  • Graph RAG
  • Cost

  Read on engineering.fb.comDetailsSave
• Meta28 Oct 2026Ads

  Meta Andromeda: Powering the Next Generation of Ads Retrieval Through Model-System Co-Design

  Problem Retrieving relevant ads from billions of candidates was too slow and expensive for high‑quality models.

  Solution Meta co‑designed a new model, system architecture, and custom GPU hardware to run a much larger neural network with hierarchical indexing and GPU‑native operators, achieving near‑ranking quality at retrieval speed.

  tens of millions ads/second per GPU

  • GPUs
  • Cost
  • Latency

  Read on engineering.fb.comDetailsSave
• Pinterest17 Sep 2026Ads

  Beyond Two Towers: Launching the 3-Tower Engagement Co-Train Model (Part 2)

  Problem Ads needed more accurate click predictions without slowing down serving millions of candidates.

  Solution Created a 3‑tower co‑train model that shares embeddings, gives a fast two‑tower score for all ads and a richer three‑tower score for a selected subset, keeping latency low.

  • Two-tower
  • Cost
  • Latency

  Read on medium.comDetailsSave
• Etsy20 Mar 2026Ads

  Making Ads Count: Using MMoE and Auxiliary Tasks to Better Connect Buyers & Sellers

  Problem Buyers struggled to find the right items and sellers couldn’t reach interested customers.

  Solution They added a Multigate Mixture of Experts architecture and used add‑to‑cart events as an extra training signal to better predict purchase intent and rank ads.

  • Reranking
  • Hybrid search

  Read on etsy.comDetailsSave