Head of Technical Education at Anthropic, teaching AI application development and coding workflows.[2]
Start with MCP: Build Rich-Context AI Apps with Anthropic →
Education
Links
LinkedIn First-person teaching interviewHead of Technical Education · Anthropic
Career
Schoppik says he graduated in finance in 2010 and learned programming while building an education business with a developer friend. He later co-founded Rithm School with Tim Garcia and Matt Lane, emphasizing smaller classes and mentorship. His teaching at Anthropic continues that practical approach through courses where learners build working applications and inspect the steps involved.[3][2]
Contributions
Learning an agentic coding workflow
His Claude Code course works through modifying existing applications, refactoring a notebook and building interfaces. The exercises include connecting tools through MCP and testing browser behavior, so coding assistance is treated as a workflow around a codebase rather than a single generated snippet. Learners see how instructions, project context and tool access change what the assistant can do.[1]
Giving an application tools and context
His MCP course starts with a chatbot that searches academic papers, then separates the external capabilities into a server. MCP defines a client-server connection through which an application discovers tools, resources and prompt templates. Learners build both sides, inspect the server and connect it to existing applications, making the boundary between model and external service explicit.[2]
Tools, writing & talks
Writing, research & talks
Start here
MCP: Build Rich-Context AI Apps with Anthropicdeeplearning.aiA hands-on course implementing an MCP server and client, connecting a chatbot to tools and deploying the server.
Start with the MCP course if you know Python, LLM prompting and application development. Build the sample chatbot and server to see which capabilities the protocol exposes.
Instructor[2]A course with Anthropic, taught by Elie Schoppik, on using Claude Code past the basics: coordinating multiple subagents on different parts of a codebase, autonomous issue-to-merge GitHub workflows, notebook-to-dashboard cleanup, and Playwright over MCP for visual debugging.
Taught by Elie SchoppikOriginal source- Claude Code: A Highly Agentic Coding Assistantdeeplearning.ai
A practical course working with existing code, interfaces and external development tools.
Instructor[1]