MCP Tools as Your Database Engineer
summary
Claims automation saves 20% development time by using AI agents to design schema, provision branches, seed data, run migrations and create PRs. Live demo uses Neon, Gibson AI, GitHub MCP servers, GitHub Copilot to create a new Postgres branch, add an address column, update backend and frontend code, push a branch and open a PR; most steps complete end to end in minutes though a few agent actions falter during the live demo.
AI-written summary · about
chapters (19)
- 0:00MCP Tools and AI in DB engineering
- 4:03DB migration pain points
- 10:03Traditional vs AI-driven DB changes
- 12:55AI workflow with DB branches
- 14:41Demo setup: travel agency
- 18:14Project setup and MCP config
- 20:26AI prompts for schema changes
- 21:11MCP tools with Neon Postgres
- 22:17Automate branch and schema updates
- 23:44AI-assisted app code updates
- 25:10AI frontend/backend changes
- 26:37PR with AI-generated changes
- 27:42Local testing of updates
- 28:49AI code review with Copilot
- 30:12Postgres MCP extension
- 31:16Chat with DB using AI agents
- 32:01AI for legacy code and migrations
- 34:15Summary and future of AI in DB
- 36:45Q&A
related talks
30mPrompt to Performance Speeding Up Database Administration with AI and Claude Code
14mReal world MCPs in GitHub Copilot Agent Mode
30mComposable AI: Building Next-Gen AI Agents with MCP
45mAgents, Tools, and MCP, oh my! Next-level AI concepts for developers
36mPutting AI Agents to Work: A New Era of Open Source Connectivity with MCP
27mPanel: APIs for AIs
21mBuilding Agents (the hard parts!)
46mPanel: AI in the Loop - Left-Shifting Platform DX and Governance