Chapter

Production ML at scale

Feature engineering, candidate generation and ranking, model training, distributed and low-latency serving, and online A/B testing, at 150M+ user scale.

Major contributor

Recommendation engine for the TikTok Shop landing page

Recommendation engine that ranks which products a shopper sees first. +129% GMV.

Major contributor

Production-scale ranking and recall models

Deep-learning ranking and retrieval models that order products at scale. +96% orders.

Major contributor

Pre-filter ranking layer above the main ranker

Ranking layer that pre-screens a large product pool so the strongest matches reach the main model first.

Sole author

End-to-end CTR prediction for push notifications

Click-through model, owned end to end, that decides which notification to send and when. +153% engagement.

Sole author

Push-notification audience targeting at 150M+ scale

Targeting model that predicts what 150M+ users want next and matches notifications to it.

Sole author

Income-proxy model for credit-card eligibility

Scoring model that estimates user income from behavior to power credit-card approvals.