HomeUse Cases → Feature Store

Real-Time
Feature Store

Sub-millisecond feature serving for ML models. Batch pipelines, streaming updates, and production monitoring - all built on Valkey Hashes.

ML InferenceFeature ServingValkey HashesStreamingProduction ML

Cookbooks

6 guides from basic feature writes to production monitoring and versioning

Live Demo

Create entities, write features, fetch feature vectors, and measure latency in real-time

How Valkey Powers Feature Stores

WRITE FEATURES
HSET fs:v1:user:user_123
age "28"
ltv "1250.5"
segment "premium"
EXPIRE key 3600

One Hash per entity. Atomic write + TTL. ~0.2ms.

READ FEATURES
HGETALL fs:v1:user:user_123
-- or select fields --
HMGET key age ltv segment

Full vector or selective fields. ~0.1ms.

BATCH PIPELINE
PIPELINE:
HGETALL user_001
HGETALL user_002
... x100
EXECUTE → 1 round-trip

100 entities in ~0.3ms. One TCP round-trip.

STREAMING UPDATES
XADD fs:stream:user *
entity_id user_123
features {json}
XREAD ... → HSET

Publish updates via Streams, materialize to Hashes.

Complete source code on GitHub

Full Python library with Entity, FeatureView, OnlineStore, StreamingWriter, and FeatureMonitor. Docker compose included.