LangGraph. Stateful agents that ship.
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 multi-step LangGraph agents in production across these businesses every week — the state-machine patterns on this page are the ones we use, not the ones from 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. Once production agents get past two tool calls, the state-management problem becomes the entire problem — and that is where LangGraph earns its place. Explicit state machines, debuggable nodes, checkpointing, replay, human-in-the-loop interrupts, multi-actor flows. LangChain core had its share of valid criticism; LangGraph is a different story. You leave the Academy with a deployed LangGraph system in production — or as the founder of an agent-shaped product whose moat is the graph, not the prompt.
Coding · LangGraph at production fluency
Explicit state graphs you can reason about, checkpoints that survive a process restart, streaming the front end can consume, long-running flows that pause for human input, trajectory evals across the whole graph — not just individual nodes. LangGraph Studio and LangSmith alongside open-source observability. A deployed multi-step agent by week twelve — the artefact that wins interviews at every AI-native team that has graduated past one-shot calls.
Commercials · agent orchestration as a productised offer
Most SMBs trying to ship agents are stuck in a v0-era LangChain prototype that no-one can debug. A “LangGraph migration” engagement at £15-30k is the new high-ticket retainer at AI-native shops. Pricing, the discovery call, a one-page pilot agreement, the first paying customer. A paid pilot by week six.
Investment · raising on agent infrastructure
LangChain ($200m+ valuation, $25m raised), CrewAI funded by a16z, AutoGen (now Magnetic-One) shipping from Microsoft, Inworld $50m, Mastra raising on the agent-orchestration thesis, Sierra ($4.5bn), Adept, /dev/agents — agent infrastructure is one of the most heavily funded categories in AI. Cap table, ten-slide deck, financial model. A live investor pipeline by demo day.
Common questions.
Is LangGraph the right answer for every agent system?
No — one-shot calls or simple tool loops should usually use provider-native APIs directly. We teach the judgement on when LangGraph earns its place; the wider LangChain page covers the comparison honestly.
Do I need LangChain core experience first?
No. LangGraph stands on its own; many teams use it without ever importing LangChain core. We will get you fluent inside two weeks.
Will I learn LangGraph Studio and LangSmith?
Yes — both, alongside the open-source agent observability alternatives. We teach you to evaluate the tooling, not lock you into one vendor’s stack.
What about multi-agent systems?
Covered — LangGraph is one of the strongest multi-agent orchestration frameworks in the current ecosystem. The agents course page is the wider framing if multi-agent is your specific entry point.
What language do you teach it in?
Python and TypeScript — both first-class in LangGraph. You will work in whichever the project demands; the patterns transfer cleanly.
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 what graph you would build and what state it would carry. James and Louis read every application personally and reply inside the week.
© 2026 Moonlabs Incubator. All rights reserved.