Learn the Model Context Protocol.
Moonlabs is the operator-led AI Academy in Derby. We run three live companies — Homemove, home.co.uk and homedata.co.uk — and we teach twelve students per cohort to ship a real AI product, sell it to a real customer, and raise on it. Three pillars: Coding, Commercials, Investment. Twelve weeks. £6,000.
Moonlabs is what we are. Two operators — James Freestone and Louis O’Connell-Bristow — who run Homemove, home.co.uk and homedata.co.uk. We run MCP servers in production across all three businesses — the patterns on this page are the ones we use to ship, not pages out of a tutorial.
The Academy is what we do. A twelve-week, in-person, twelve-student cohort in Derby. You build a real AI product. You sign a paid pilot on it. You write a deck and a financial model. You leave with a deployed system, a paying customer reference and a live investor pipeline. Coding, Commercials, Investment — the three pillars taught in equal weight every week.
Why this page exists. The Model Context Protocol is the part of the agentic AI stack that quietly became inevitable in 2025. Claude Code, Cursor, Codex and most serious AI tooling now speak it. Almost no course teaches it properly — the docs are getting there, the tutorials lag, and the people who use it daily are mostly engineers inside AI-native teams. You leave the Academy with your own MCP servers running in production — or as the founder of an MCP-first product on top of the protocol nine months before most of the market notices.
Coding · production MCP, end to end
Production MCP servers in TypeScript and Python. Tool scoping, resource discovery, authentication, transport, error handling, evals for tool calls in CI, observability across the boundary. Then write the clients and wire both into Claude Code, Cursor and Codex. A deployed MCP server by week twelve — the artefact that lands interviews at every AI-native team in the country.
Commercials · MCP as a productised offer
MCP servers are the new “Zapier integration” line item. Every B2B SaaS now needs one; almost none of them have engineering capacity to build it. Pricing per-server retainers, the discovery call, a one-page pilot agreement, the first paying customer. A paid pilot by week six — for many graduates, a productised MCP-server offer sold into their existing network.
Investment · raising on MCP-first products
Anthropic shipped MCP and Cursor/Codex backed it inside weeks — the protocol-momentum compounds every quarter. Pulumi, Apify, Toolhouse raised on AI tool infrastructure; ChatPRD, Mem, Granola all shipping MCP-first surfaces; Cline, OpenInterpreter, Goose getting funded by Tier-1 funds. Cap table, ten-slide deck, financial model. A live investor pipeline by demo day.
Common questions.
Do I need experience with MCP before applying?
No. We will get you fluent inside two weeks. MCP itself has the shortest learning curve of any protocol we have taught.
What languages will I build servers in?
TypeScript and Python are the defaults — both are first-class in the MCP ecosystem. Other stacks (Go, Rust, PHP) appear where projects demand them.
Will I learn MCP clients as well as servers?
Yes — both sides. You will write servers, then write clients that consume them, then wire both into the agents (Claude Code, Cursor, Codex) you already use. The full loop, not just the easy half.
How does this fit with the wider Academy curriculum?
MCP is one piece of the broader LLM engineering surface. The LLM engineering page is the wider framing; the agents course page is the sibling framing for the agent-first entry point.
Is this an Anthropic-sponsored course?
No. The Moonlabs Academy is independent. We teach MCP because it has become the de facto standard for tool integration with serious AI agents — not because Anthropic asked us to.
More Academy entry points.
The Academy is one course with many doors. Each of these pages is a different entry point into the same twelve weeks.
Build it. Sell it. Raise on it. In twelve weeks.
Tell us where you are starting from and what you would wire an MCP server into first. James and Louis read every application personally and reply inside the week.
© 2026 Moonlabs Incubator. All rights reserved.