Anatomy of a Self-Service Analytics AI Agent
How we built a multi-agent system that turns natural language questions into Databricks SQL and Power BI DAX: what works, what doesn't, and what's next.
Read article →Data · AI · Business Intelligence
I write about what happens to BI when AI starts writing the queries.
I lead BI and data work day to day, and here I write about metric definitions, semantic layers, guardrails and the practical side of letting AI into analytical workflows.
Current focus
AI-Augmented Analytics
How AI changes the way analytical products, dashboards and decision-support tools are created.
Metadata & Semantic Quality
Why reliable AI-assisted analytics depends on clear definitions, governed metrics and trustworthy semantic layers.
Analytics Leadership
Practical reflections on BI modernization, platform thinking, data teams and the changing role of analytics professionals.
Writing
Latest notes and essays
AI-Assisted Investment Analytics: Why the First Answer Was Not Good Enough
A case study on AI boom stocks, 10-K disclosures, factor models, placebo tests, and the role of human judgment in investment analytics.
Read article →
Python as the Glue Layer for AI-Augmented Analytics
A Streamlit demo showing how Python can connect business questions, metadata, AI-generated SQL, validation and analytical explanation in one lightweight workflow.
Read article →Featured project
Python + AI as a practical analytics prototype layer
A Streamlit-based presentation app showing how Python can connect business questions, data, AI-assisted reasoning, SQL-like logic, validation and explanation in one lightweight workflow.
Open demo