KK kedro-datasets 9.2.0 is now live! KK This release brings enhanced tracing support for agent workflows, improved API handling, and new experimental MLRun datasets. 🎉 🚀 Major features and improvements • Added

autogen

mode to

LangfuseTraceDataset

and

OpikTraceDataset

for tracing AutoGen agent conversations with OpenTelemetry integration. •

api.APIDataset

now stores the response received from a

PUT

or

POST

request via the

response_dataset

parameter. • Experimental

OpikTraceDataset

and

LangfuseTraceDataset

now receive OpenAI credentials as

base_url

and

api_key

(instead of

openai_api_base

and

openai_api_key

). 🧪 New experimental MLRun datasets We’ve introduced a new set of experimental MLRun datasets under `kedro_datasets_experimental.mlrun`: •

MLRunAbstractDataset

— Base dataset for MLRun integration, usable for generic artifacts; •

MLRunModel

— Save and load ML models via MLRun with framework metadata; •

MLRunDataframeDataset

— Save and load pandas DataFrames as MLRun artifacts; •

MLRunResult

— Log scalar results and metrics to MLRun;

Note: These MLRun datasets are experimental and may evolve in future releases. We’d love your feedback if you try them out!

🐛 Bug fixes and other changes • Fixed

ibis.TableDataset.exists()

to correctly account for the

database

(schema/collection of tables). 🙌 Community contributions Thanks to our contributor: @Ekaterina Molchanova and Iguazio team! We welcome all contributions — big or small. Try out the new features, share feedback, and help shape the next release! Until next time, The Kedro Team 💙