faust-streaming/faust

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

1884 stars · 205 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Faust-streaming is a Python stream processing library that ports Kafka Streams ideas to Python using asyncio, letting developers build distributed real-time data pipelines and event processors with plain Python code. It is a community-maintained fork of the original Robinhood Faust project with continued releases, Kafka transaction support, and updated aiokafka drivers.

Use cases

• process kafka streams in python
• build real-time event processing pipelines
• consume and transform infinite event streams
• maintain durable in-memory key/value tables from event streams
• implement windowed aggregations like click counts
• build high-performance distributed systems with asyncio
• process billions of events per day in python

When to choose

• you want Kafka Streams-style processing in pure Python without a DSL
• you need to use Python libraries like NumPy, Pandas, or PyTorch inside stream processors
• you want asyncio-native agents consuming from Kafka topics
• you need persistent, windowed state tables backed by RocksDB

When to avoid

• you need heavy-duty exactly-once semantics at massive scale where Flink or Spark Streaming are proven
• your team prefers JVM-based stream processing ecosystems
• you don't already operate a Kafka (or compatible) broker
• you need a maintained fork guarantee — the original project's release process was the reason for this fork

Facets

library · maturity active

streaming message-queue serialization etl microservices big-data python kafka asyncio stream-processing kafka-streams rocksdb event-processing redis data-engineering real-time linux macos docker kubernetes

Member repositories

For agents

markdown · JSON · MCP: product_card(name="faust-streaming/faust")