AI Studio - Creating and scaling your custom copilots
summary
Azure AI Studio is now generally available and claims to let teams ship multi‑agent copilots end‑to‑end: the demo builds a Contoso sales copilot that uses the Assistant API plus a function that auto‑writes SQL against a sales DB and the code interpreter to generate charts. Prompt Flow adds tracing (OpenTelemetry spans showing the LM prompt, SQL and generated Python), evaluations (custom SQL similarity, error rate, safety checks across models) and monitoring to App Insights. Real numbers shown: 20,000 monthly AI Studio users, 53,000 orgs on Azure AI, and model comparisons where GPT‑4 variants trade speed for best SQL accuracy.
AI-written summary · about
chapters (10)
- 0:00The state of AI
- 2:41Azure AI studio launch
- 3:38Customer success with AI
- 5:00AI studio platform tour
- 10:48Building custom co-pilots
- 13:12Sales insights application
- 16:11Introducing Prompt Flow
- 18:27Tracing and debugging
- 25:57Model evaluation
- 33:53Monitoring and future outlook
related talks
15mAI Studio - Creating and scaling your custom copilots
45mWelcome to the world of SQL Copilots
45mThe era of the AI Copilot
58mNext generation AI for developers with the Cloud
10mAI Co-Pilot Demo
55mBuilding and scaling AI applications for Data Scientists and Developers
01h 58mBuild, Evaluate and Deploy a RAG-Based Retail Copilot with Azure AI- 45mWelcome to the world of SQL Copilots