Chronon
Chronon is a data platform for serving for AI/ML applications. It is an open source end-to-end feature platform that allows Machine Learning (ML) teams to easily build, deploy, manage and monitor data pipelines for machine learning.
Key Features
- Consume data from a variety of Sources – Event streams, batch warehouse tables, production databases, change data streams, service endpoints, as well as data available in the request context can all be used as Chronon inputs.
- Easily define transformations and aggregations – Raw data ingested from sources can be transformed into useful features using a flexible library.
- Produce results both online and offline contexts – Online, as scalable low-latency end-points for feature serving, or offline as hive tables to power model training and evaluation flows.
- Real-time and batch support with online/offline consistency – Chronon not only supports realtime and batch updates to the online KV store used to power production inference, but it also provides temporally accurate backfills in the offline context. This means that training data is always consistent with production inference data.
- Powerful python API – Define your features in a simple and intuitive python interface, and let Chronon abstract away the complexity of computation and serving.
- Automated feature monitoring – Auto-generate monitoring pipelines to understand training data quality, measure training-serving skew and monitor feature drift.
Resources:
Chronon — A Declarative Feature Engineering Framework (blog post)
Chronon, Airbnb’s ML Feature Platform, Is Now Open Source (blog post)