Authentic Intelligence Labs is an open-source initiative dedicated to solving the "Definition Drift" problem in modern data stacks. We build vendor-neutral protocols that decouple business logic from BI tools, enabling a true Headless Data Governance architecture.
In the modern data stack, business logic is often fragmented. A "Gross Margin" calculation in dbt might conflict with the DAX formula in Power BI, which differs again from Tableau.
We believe that:
- Definitions should be defined once, not reinvented in every tool.
- Governance should be "Headless"—managed as code (JSON/YAML) and synced downstream.
- Standards should be open, not locked into a proprietary catalog.
Our flagship project provides the schemas and validators needed to implement the Open Governance Standard.
The vendor-neutral JSON standard for Headless Data Governance.
What it does:
- Acts as the "API" for your business metrics.
- Defines a single source of truth for KPIs, Data Quality rules, and Lineage.
- Syncs definitions automatically to dbt, Power BI, Tableau, and Data Catalogs.
Core Components:
standard_metrics.json: The "Golden Record" for KPIs.standard_dq_dimensions.json: 60+ industry-standard data quality dimensions.standard_data_rules.json: Technical validation rules (Regex, null checks).root_cause_factors.json: Standardized taxonomy for data incidents.
We don't just define standards; we demonstrate how they work in the real world.
- Write Once, Sync Everywhere: See how a single JSON change triggers updates across your entire stack.
- CI/CD for Governance: Validate your metrics with Python-based validators before they break your dashboards.
🌐 Live Demo: metricprovenance.com
We are an open community and welcome contributions from Data Engineers, Analytics Engineers, and Governance Leads.
- 🐛 Found a bug? Open an issue in our main repository.
- 💡 Have an idea? Submit a Pull Request to expand the OGS schema.
- 💬 Discuss: Connect with us to shape the future of Headless BI Discussions.
