IBM Db2 Genius Code brings AI-assisted SQL authoring, troubleshooting and optimization directly into Visual Studio Code, grounded in the developer’s actual database schema to help generate, understand, troubleshoot and optimize SQL while keeping recommendations transparent and developers in control. IBM said general-purpose AI assistants might not understand an organization’s Db2 environment, database schema or governance requirements. With IBM Db2 Genius Code, developers can generate SQL using natural language through the IBM Db2 Genius Code chat, refine queries conversationally while retaining context across multiple interactions, understand complex SQL through plain-language explanations, optimize queries with AI-assisted recommendations and annotated explain plans, review performance and resource usage before executing a query, and create a library of verified queries that can be reliably reused. IBM Db2 Genius Code connects AI-assisted query refinement with performance analysis; its performance analyzer capabilities can measure changes in CPU consumption, input and output activity, and latency. AI-generated suggestions remain editable, recommendations can explain their reasoning, and queries don’t run without developer approval. Built-in guardrails and governance help organizations support secured development through governed data handling and reviewable outputs.