The Future is Non-Deterministic
summary
Claims that non‑deterministic problem solving is practical now through agentic workflows using tool calling and shows how that works end to end: a standard Model Context Protocol ties LLMs to isolated mCP servers, Docker wraps those servers as containers and publishes them to Docker Hub via the Docker mCP catalog, and live demos use Claude and Cursor to call a YouTube transcript tool and a Notion mCP server. Demos sometimes hit parameter errors but recover after prompt or code tweaks and the container sandboxing enforces a clearer security model.
AI-written summary · about
chapters (6)
- 0:00Introduction and background
- 1:14The evolution of AI agents
- 5:55Model context protocol
- 7:46Docker and MCP integration
- 9:58Live demo and use cases
- 25:39Recap and final thoughts
related talks
36mComposable Architecture for AI Systems: APIs, MCP, Skills, Sub-Agents, and Wor…
16mDriving agentic innovation w/ MCP as the backbone of tool-aware AI
34mLLMs, Meet Your Tools: The Magic of Model Context Protocol
29mFrom Copy-Paste to Composition: Building Agents Like Real Software
01h 00mFrom LLM Theory to Practical Agentic Implementations
07mJust do it. (let your tools think for themselves)
45mModel Context Protocol (MCP): The future of user interactions?
19mThe Future of MCP