Where most of my time goes: a family of open source products bringing AI agents into analytics engineering across Databricks, Microsoft Fabric, and Power BI — three verbs,understand → operate → consume. In production with enterprise data teams, built on a simple idea: AI as a colleague, not a replacement.
ade-catalog
Understand — what exists, and how it’s connected
A cross-platform analytics metadata catalog. It extracts metadata from Databricks, Power BI, and Fabric, builds the cross-platform lineage native tools don't, and exposes it through a web UI and an MCP server for AI agents.
Open sourceGitHub →ade-ops
Operate — one source of truth, safe promotion
An operations framework for analytics teams across Databricks, Fabric, and Power BI. One source of truth, multi-environment promotion, and diff-before-push safety — designed for humans and AI agents to operate together.
Open sourceGitHub →ade-lens
Consume — the certified answer, with the proof
A governed, cross-tool consumption layer over an existing BI estate (Power BI, Databricks, legacy BI). It doesn't hand you a plausible answer — it takes you to the certified report, with the proof.
LiveVisit →