logistics_ops · workflow
DoorDash builds an ML Platform to standardize and scale machine learning across logistics and marketplace
As ML usage grew across DoorDash—spanning fraud prediction, search ranking, delivery time predictions, and recommendations—there was no holistic ML Platform to standardize frameworks, manage model lifecycles, or compute and serve features at scale, limiting engineering and data science productivity.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Real-time event ingestion
A Realtime Feature Aggregator listens to a stream of events and aggregates them into features stored in the Feature Store.
Tools used
LightGBMPyTorchTorchScript
Outcome
DoorDash standardized on LightGBM for tree-based models and PyTorch for neural network models, and designed a four-pillar ML Platform architecture—Modeling Library, Model Training Pipeline, Features Service, and Prediction Service—to support end-to-end model lifecycle management at scale.
Results
Volumeaccuracy of models were almost the same
Grounding & classification
Source type: technical build writeup
18 fields verified against source quotes, 4 dropped as unverifiable.
fraud detectionpredictive analyticsrecommendation systemproduct catalogsource backedtools describedworkflow describedecommercelogisticsemployee productivitytechnical build writeupback office opslogistics opsdata sync enrichment