ecommerce_ops · workflow
Instacart supercharges ML/AI foundations with Griffin, Feature Store, and Axon in H1 2023
ML models at Instacart were trained on laptops or siloed team infrastructure with no common patterns, and promoting a model to production often took more than a month, limiting ML velocity and sophistication across the company.
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 · User Modeling signal generation
User Modeling generates user-level signals that enable data scientists and MLEs to characterize customers across multiple behavioral dimensions.
Tools used
GriffinFeature StoreAxonFeature MarketplaceTensorFlowTensorFlow ServingRayLightGBMSageMaker
Outcome
Instacart reduced feature onboarding from days to under an hour, achieved up to 85% faster queries, cut TensorFlow P99 latency by 85%, enabled training with 10x or even 100x more data, and halved Feature Store ingestion costs — substantially accelerating ML productivity across the company.
What failed first
The only GPU access path for MLEs was AWS SageMaker, which required multiple rounds of terraform configuration that could take days to weeks for first-time provisioning and was difficult to maintain for dependency management.
Results
Time savedless than an hour (vs. a few days previously)
Volume85%
Cost replacedcut its ingestion costs in half
Running sinceearly 2021
Grounding & classification
Source type: technical build writeup
40 fields verified against source quotes.
forecastingfraud detectionpersonalizationpredictive analyticsrecommendation systemknowledge basebuilder submittedfailure mode describedmetric backednamed customerproduction runtime claimedsource backedtools describedworkflow describedecommercecost reductioncycle time reductionemployee productivityresponse time reductionthroughput increasetechnical build writeupback office opsecommerce opsdata sync enrichment