Free AI agent projects, built live every Thursday.
Each Live Lab is a sixty minute Lightning lesson on Maven. Param builds one agent project from a public repo, answers questions, and the code lands on GitHub for you to run on your own work.
Save a free seatEvery lab ships to its own folder in agentic-ai-lightning-sessions on GitHub.
October
Build a Mission Control for AI Agents
A board of AI employees that resolve a company's billing issues, where a refund waits for your tap and a card closes only when the payments ledger agrees.
Save a free seatBuild a Multi-Agent System with MCP
A planner, an engineer and a reviewer that take one issue from plan to pull request, each with only the tools its role needs.
Save a free seatNovember
Secure AI Agents with Guardrails
Agents that each sign in with their own identity, act only within the user's permissions and refuse an injected instruction.
Save a free seatCut AI Agent Costs with Token Economics
One agent run priced step by step, rolled up to a cost per feature and per customer, and cut with routing, caching and a budget.
Save a free seatBuild a Second Brain with Claude Code
A second brain over your notes, docs and Linear that answers with sources and serves the same memory to any agent over MCP.
Save a free seatDecember
Build AI Agent Evals from Production Traces
A test set built from real agent runs, scored by code checks and an LLM judge, that fails the build when the agent gets worse.
Save a free seatTrace AI Agents in Production
Every agent run traced in Langfuse, a dashboard of cost, latency and failures per agent and per customer, and an alert before one goes over budget.
Save a free seatBuild an AI Agent Governance Framework
A way to decide which work goes to agents, who approves what they do, and what they changed each month.
Save a free seat
Why do agent side projects stay side projects?
A weekend project teaches the API. Work asks for what the weekend skipped.
The project has no real users. It never meets permissions, approvals or a second tenant, so it never teaches them.
The cost is invisible until the bill arrives. Nobody measured tokens per outcome while it was a toy.
It breaks the same way twice. Without a trace you guess at the cause and patch the prompt.
Each lab starts from one of those production problems and builds the fix on a public repo, so the project you clone already has the part a weekend skips.
What a Live Lab gives you
A project worth showing.
Each lab ends in a repo folder you can fork, extend and put on your profile.
One production skill at a time.
Control, security, cost, traces, evals and governance each have a lab of their own.
Questions answered live.
The last part of each lab is open questions with Param.
A recording if you miss it.
The full recording follows here, free with an account, and its video course on /courses.
One system a week, built where you can read every line.
From seat to running code in three steps
Thursdays at 16:00 UTC in October, 17:00 UTC from 5 November.
Save a free seat
Sign up on Maven with your email.
Build it live
The problem, then the build, then questions.
Run it on your code
Run it on your code and post what broke in Discussions on the live-labs repo.
Watch a past lab first
Questions about the Live Labs
Are they really free?
Yes. Lightning lessons are free on Maven. You sign up with your email, and the recording stays on Maven after the session.
What level do I need?
You can read and run a Python or TypeScript repo, and you have called an LLM API before.
Who are they not for?
People who want a slide talk with no code, or who are new to programming.
Do I need a paid model account?
Bring your own API key for whichever model the lab uses. Each repo shows where to swap the model.
Can I ask about my own project?
Yes, in the questions part of the lab or in Discussions on the live-labs repo.
How do labs relate to the workshops and the buildcamp?
A lab builds one piece in an hour. A workshop builds a whole slice in four hours. The buildcamp builds two production systems over four weeks.
What is settled before you save a seat
The lab costs nothing and needs only your email.
The recording stays on Maven.
The code is public and yours to fork.



