Workshop Flow
Step 1: Deploy solution
Deploy infrastructure and run with pre-built sample scenario:
- Deploy Microsoft Foundry and Azure resources (AI Services, AI Search, Storage)
- Configure Microsoft Fabric connection
- Configure your development environment
- See the agent working with sample data
- Takes ~15 minutes
Step 2: Customize for your use case
Generate custom data for each use case:
| Customer Industry | Use Case Example | Sample Questions |
|---|---|---|
| Telecommunications | Network outages + service policies | "Which outages exceeded our SLA threshold?" |
| Manufacturing | Equipment data + maintenance docs | "Which machines are overdue for maintenance per our schedule?" |
| Retail | Product catalog + return policies | "What's our return policy for electronics over $500?" |
| Finance | Account data + lending policies | "Which loan applications meet our approval criteria?" |
| Insurance | Claims data + policy documents | "What's the status of claims filed this week vs our SLA?" |
| Energy | Grid monitoring + safety protocols | "Which substations are operating above 80% capacity?" |
| Customer X | Their data + Their docs | Their burning questions |
Pre-PoC prep
Run Step 2 before your PoC. Enter the industry and a brief use case description. The AI generates realistic sample data, documents, and test questions tailored to your scenario.
Step 3: Deep dive
Prepare for technical questions in customer conversations:
- Fabric IQ: How ontology translates business questions to SQL
- Foundry IQ: How agentic retrieval plans, iterates, and reflects
- Orchestrator Agent: How the agent decides which source to query