ADVANCED
Conversational Analytics Agent: Semantic Layer to Production
Build a semantic layer over realistic SQL business data, then a Claude-powered agent that answers KPI questions — and deploy it to a URL.
LiveAdvanced 8 weeksreading + build-along labs + capstone
8 modules29 lessons7 problem sets1 capstone
Your progress0%
What you'll learn
- Explain why raw SQL / text-to-SQL fails for business metrics
- Model a business as entities, dimensions, and governed metrics
- Run, verify, and extend a semantic layer that compiles analytics requests into correct SQL
- Test metrics for correctness and avoid fan-out bugs
- Build a Claude tool-use agent that never writes SQL
- Guard against hallucinated numbers and evaluate the agent
- Deploy the agent to a public URL with a secure API key
Curriculum
Every lesson below is a full, written lesson — click any of them to start.
Certification
Complete all modules, pass problem sets, and finish the capstone to earn:
Certified AI Analytics Engineer
- ✓ Builds a governed semantic layer over raw SQL data
- ✓ Designs metrics and dimensions that are correct by construction
- ✓ Builds a Claude tool-use agent that answers business questions safely
- ✓ Deploys a conversational analytics app to production
Credential format: EDOVA-CAA-2026-XXXX
Edova Academy
Certified AI Analytics Engineer
Awarded to
Your Name
Credential ID: EDOVA-CAA-2026-DEMO
Verified · Sample credential