Bruin AI Step 2 of 4

Build a Self-Healing Pipeline Agent

Choose the first pipeline

Start with one production pipeline that has failed before. Do not document the whole warehouse on the first pass.

What to do

  • Pick a pipeline with a recent failed run and a clear business output.
  • Pick one output asset in that pipeline, such as analytics.daily_orders.
  • Run the command below from the repository root. Replace the path with the asset file you picked.
bruin lineage --full assets/analytics/daily_orders.sql
  • Write down its upstream and downstream assets, data owner, and the people affected by bad data.

Why the agent needs this: it shows the agent which output failed, what feeds it, and what else could be affected.

Sign up to our newsletter

Practical updates on open-source data pipelines, AI analysts, governance, and what we are shipping at Bruin.

The signup form is hosted by Brevo. Accept cookies to load it.