How do I do agentic data engineering with an AI agent?
Through Bruin MCP an agent can build and maintain pipelines against governed assets: it sees lineage and checks, writes SQL and Python assets, and runs them, all within the guardrails you define.
Command
bruin mcpDefined in
MCP + YAML
Works with
Any MCP client + Bruin
What you get
Agent-authored assetsgoverned guardrailslineage-aware
How it works in code
{
"mcpServers": {
"bruin": {"command": "bruin", "args": ["mcp"]}
}
}Run bruin mcp and the agent builds and maintains pipelines within your guardrails.
Give agents governed data
Open source. Expose assets, quality, and lineage to AI agents over MCP.