Flyte, an LF AI & Data Foundation Incubation Project, has released version 0.13.0! Flyte is a Kubernetes native workflow automation platform for complex, mission-critical data and ML processes at scale. It allows users to describe their ML/Data pipelines using Python, Java, Scala, or in the future, other languages and manages the data flow, parallelization, scaling, and orchestration of these pipelines.
In version 0.13.0, Flyte adds a variety of improvements. Highlights include:
- Support for complete Oauth2 spec including SingleSignOn and configuration examples for popular IDP’s now available in Flyte. Please see the updated information and description of the feature, and the setup information.
- Backend improvements to support dynamic workflow visualization and simpler debugging for Kubernetes and external errors (UI updates in future releases).
- flytectl: Cross-platform portable CLI for Flyte
- Documentation site overhaul and redesign
- Improved end to end platform performance
- Beta: Updated API to interact with past executions, launch new executions. This makes it possible to have simplified programmatic access for all Flyte features and perform regular data science tasks like retrieving previous results, compare executions, etc.
- Beta: Support for prebuilt container plugins with faster user interactivity
- Plugin: Interact with any SQL database using SqlAlchemy
- Plugin: Use Versioned datasets using DoltDB
- Access secrets using standardized interaction pattern
To learn more about the Flyte 0.13.0 release, check out the full release notes. Want to get involved with Flyte? Be sure to join the Flyte-Announce and Flyte-Technical-Discuss mailing lists to join the community and stay connected on the latest updates.
Congratulations to the Flyte team and we look forward to continued growth and success as part of the LF AI & Data Foundation! To learn about hosting an open source project with us, visit the LF AI & Data Foundation website.
Flyte Key Links
LF AI & Data Resources