Jarvis AI Agent for Database Migration | IBM Db2 to Azure Cosmos DB
This video walks through a practical database modernization flow where a Jarvis Registry agent converts an IBM Db2 schema into an Azure Cosmos DB-ready model. Instead of treating migration as a direct lift-and-shift, the demo shows how teams can rethink schema structure, relationships, partitioning, and performance constraints in a guided process that is easier to scale than manual redesign.
The agent runs directly in the IDE, reviews source tables, columns, keys, and relationships, then asks follow-up questions about real access patterns and data usage. That back-and-forth helps shape a target design aligned to application behavior rather than assumptions, giving developers real-time feedback before conversion begins and reducing rework later in cloud migration projects.
After user approval, the agent executes conversion and produces a Cosmos DB model that can be reviewed immediately in Azure, including containers, document structure, and partition design. The demo highlights how Jarvis AI accelerates enterprise modernization with AI-assisted analysis, interactive validation, and automated schema conversion while keeping decisions transparent throughout the workflow.
The video explains why relational-to-cloud-native migration is not a simple copy process. Teams must redesign schemas, query patterns, partitioning, consistency, and performance to fit the target platform.
The presenter introduces an AI agent built on Jarvis Registry and frames the working example: converting IBM Db2 to Azure Cosmos DB. The agent runs directly in the IDE and provides real-time feedback during review and conversion.
The agent inspects tables, columns, keys, and relationships, then asks follow-up questions about access patterns and partitioning. This iterative conversation helps the design reflect how data will actually be used.
Once enough context is gathered, the agent asks for user approval of the proposed design. Users can keep refining through conversation, then trigger conversion with a single instruction when ready.
The final section shows the converted model in Azure Cosmos DB, including generated containers, document structure, and partition design. The close emphasizes faster, guided, and scalable modernization with AI-assisted conversion.
