Release time is stressful, but seeing all the hard work paying off is freaking awesome.

Lo que se conoce como dpm vaya

See you there!

www.meetup.com/openmetadata...

There are so many features you can exploit once metadata reaches the platform. The fun starts once the metadata is ingested.

You can actually activate your metadata by handling any runtime exceptions and push the pipeline status in OM as well to keep tabs on your executions and even hook it up with OM's observability system to receive alerts when pipelines fail.

Now, a follow-up question would be "all this just to document"?

We're not only talking about the importance of centralizing information, sharing context, enabling lineage, root-cause analysis,...

2. We also understand that for in-house systems, there might not be a solution already built. You can leverage the Python SDK to push pipeline and lineage information at the time the ETL itself runs. This is a very flexible approach, and this same SDK is the one that powers all connectors.

Might not be as marketing-y as clustering duckdb, but every company ends up having specific problems to solve, and we're doing an awesome job with ours.

Is anyone happy with their oob migration solution? Well, we built our own to have complete control of the execution and even allow us to migrate data using Java code directly when SQL was falling short. We also built a profiling and DQ system from the ground up, an ingestion framework,...