Andrew Ng partners with Anthropic to offer new AI developer course on MCP

- Andrew Ng Announces New AI Development Course - Providing Practical Training Using MCP \[Unblock Media] Andrew Ng, a pioneer in machine learning education, has announced a new AI development course. This short-term program called "MCP: Build Rich-Context AI Apps with Anthropic" is a hands-on curriculum that utilizes Anthropic's open protocol called MCP (Model Context Protocol) to build AI applications connected to tools, data, and prompts. The course will be led by Elie Schoppik, Head of Technical Education at Anthropic, and will involve practical implementation of MCP architecture in a client-server structure. Participants will build their own MCP server and implement functionality to connect with a chatbot that can access external data in real-time. ### Simplifying AI Integration with MCP MCP is a standardized protocol that allows LLM-based applications to communicate with external systems—such as web search APIs, local documents, GitHub code, and more. Traditionally, each use case required separate integration efforts, but MCP simplifies this by offering a single interface, significantly boosting developer productivity and flexibility. The course will especially cover the following practical exercises: Setting up a local MCP server based on FastMCP and testing with the MCP Inspector Implementing a chatbot equipped with an academic paper search tool Conducting visual tests by integrating with Claude Desktop * Practicing connections with Anthropic's open-source MCP servers (fetch, filesystem, etc.) MCP Development Roadmap: Multi-Agent, Registry, and Authentication The latter part of the course will introduce the future directions for MCP development. This includes architecture supporting collaboration among multiple agents, automatic connection based on server registry APIs, and features for server authentication and authorization management. These enhancements are expected to provide a safer and more scalable agent orchestration environment within the LLM-based AI ecosystem in the long term. Andrew Ng stated, “MCP, which enables AI applications to be more richly connected to external contexts, will become an important component of AI infrastructure moving forward.”