Data and AI releases

This month1,068Companies169

1,068 this month · 169 companies

feldera 0.365.0

UpdateyesterdayUnverifiedAdded Oct 14, 2026

- [pipeline-manager] Add resources.volume_attributes_class [adapters] Add nats_output connector for publishing to NATS JetStream Wire the Unity Catalog deletion-vector test into CI via OIDC federation

Topics: Vector search, Streaming, Governance, Data integration, Observability

Feldera's release notes

More feldera releases

Every feldera release
Updateunverified

feldera 0.364.0

- [SQL] Speed-up SQL compiler [connectors] Pass only the query string and not the full HTTP request to format parsers [adapters] Kafka: Fix a partitions list that is not in ascending order [connectors] Add connector_metadata query parameter to the http input connector [profiler] Glow around a selected or hovered expanded region [DBSP] Fix the implementation...

Updateunverified

feldera 0.363.0

- [profiler] Size an expanded region for its name and its counter [profiler] Draw the diagram from the palette-driven stylesheet [adapters] Bound Parquet row groups in the Delta writer [profiler] Aggregate PR: node text, selection mark, minimap, pressable chips, theme and browser tests [pipeline-manager] Return rustc diagnostics as structured messages...

Updateunverified

feldera 0.362.0

- [pipeline-manager] Manage deployments from deployment_config [SQL] Reject CREATE INDEX on a view with an emit_final annotation [dbsp] Make a spine's levels merge ten batches at a time [SQL] Build a best-effort parser for Rust regular expressions and use it to warn of unsupported constructs add documentation for data that is sent to Feldera

Updateunverified

feldera 0.361.0

- [SQL] Rewrite RIGHT JOINs as LEFT JOINs using Calcite [SQL] Several fixes related to ARRAY_AGG [ci] Run the Iceberg follow tests [adapters] Delta: gate two fork patches, and pin the bad-date issue [adapters] Yield after each JobQueue job.

Updateunverified

feldera 0.360.0

- [tests] Delta: assert the DV survivors' values, not just their count [dbsp] Use LZ4 as the default storage compression. [dbsp] Remove mimalloc dependency [adapters] Let the Postgres CDC connector run as a non-superuser [dbsp] Write checkpoints in the background.

Updateunverified

feldera 0.359.0

- docs: fix dead Debezium documentation links fix(s2): make pause retry test deterministic [adapters] Delta: store the ordered-snapshot resume point raw Add rust-toolchain.toml [dbsp] Properly spill batches to disk on ShardedAccumulator receive path

Updateunverified

feldera 0.358.0

- [SQL] NULL values can never be late; do not insert them in the emit_final Window operator, since they would leak memory [SQL] Fix GC for ArgMinSome (ARG_MIN with nullable argument); comprehensive MIN/MAX/ARG_MIN/ARG_MAX GC tests connector: Bump the delta-rs fork for the nested-container field-id fix [web-console] Add an "Open support bundle" button to the...

Updateunverified

feldera 0.357.0

- [adapters] Take over a checkpoint bucket at startup with take_bucket_ownership [SQL] Do not merge two GC operators if their control inputs are not 'equivalent' [SQL] Fix implementation of WATERLINE for LEFT JOIN [SQL] Fix incorrect WATERLINE computation for ASOF JOIN key (currently unreachable) [SQL] Fix HOP/TUMBLE implementation to handle dates before...

Also shipped on Oct 13, 2026

Feldera in October 2026
Betaunverified

What's included:

- Three training methods: Supervised fine-tuning (SFT) , direct preference optimization (DPO) , and reinforcement learning with group relative policy optimization (GRPO) . Each method trains a low-rank adapter (LoRA) or the full model weights. Custom loss functions : Train with any objective you compute in PyTorch from the model's log-probabilities.