How Oracle AI Database@Azure, OCI Managed MCP Servers, and Azure AI services fit together in the workflow developers already have open. 

A developer in VS Code asks GitHub Copilot a question about their data. Copilot invokes an Oracle MCP tool. SQLcl runs the query securely against Oracle AI Database. The result lands back in the IDE — no terminal-switching, no cloud-console detours, no credentials pasted into a chat window.  

Oracle AI Database@Azure sits at the heart of this workflow—enabling developers to build AI-enabled enterprise applications using Oracle AI Database services, Oracle MCP, GitHub Copilot, Azure OpenAI Service, Microsoft Foundry, and Microsoft IQ inside familiar Azure-native workflows. 

Learn more at Microsoft Build 2026

Register for Build and attend our 3 sessions, online or in person, to see these workflows in action: 

  • Build AI Apps with Oracle AI Database@Azure, MCP, and GitHub Copilot (June 3, 2:30pm PDT)  
  • Improve analytic models with Oracle Data Science Agent and Azure OpenAI (June 2, 12:20pm PDT)  
  • Move from data to intelligence with Oracle MCP and Microsoft IQ (on-demand)  

What’s new? 

SQLcl + MCP + GitHub Copilot. SQLcl now ships with native MCP support. Point GitHub Copilot at it in VS Code and Copilot can run governed SQL against Oracle Database through natural-language conversation — using the connection you already have, with the identity you already use. 
2. OCI Managed MCP Servers for Oracle AI Database. Skip the “deploy your own MCP server” step. HTTPS-based, OCI-identity-integrated MCP servers with governed toolsets, validated SQL Reports, and centralized administration. AI agents, Copilot, Microsoft Foundry, and Microsoft IQ all talk to Oracle through one managed, audited surface. 
3. Oracle AI Vector Search inside the database you already trust. Store embeddings next to your relational data, run similarity search with SQL, build RAG without standing up a separate vector store. That means grounding Copilot, Azure OpenAI, and Microsoft IQ on current enterprise data — no duplication, no sync pipeline. 
 

Why this matters for enterprise app teams 

Most “AI-native” application stacks fragment across three or four products: operational DB, vector DB, orchestration layer, agent framework. Each piece is another sync pipeline, another auth boundary, another thing to govern. 
Oracle AI Database@Azure collapses that. The data, the vectors, the MCP interface, and the governance live in one place — and they connect to the Microsoft tools your developers already open every morning: VS Code, GitHub Copilot, Azure OpenAI, Microsoft Foundry, Microsoft IQ. 
 

A low-friction starting point 

Oracle Autonomous AI Database Serverless gets you a database in minutes with Vector Search, JSON APIs, and semantic retrieval ready to use. Pair it with Azure AI services and you can have a working RAG prototype before lunch. 

Come see us at Microsoft Build 2026 

Building AI applications, modernizing Oracle workloads on Azure, or trying to figure out where MCP fits within your stack? Come to the sessions, stop by the booth to see a demo, bring hard questions. We’re looking forward to meeting you and discussing your specific requirements.