Build Your Own Coding Agent

Matt Wynne

Cucumber co-founder & agent builder

Steven Diamante

Technical coach for AI coding agents

Understand the tool you use every day

Super early bird: $450 until November 8

You use AI agents every day: Claude Code, Codex, Cursor. But do you know what's actually happening under the hood? It turns out, the core of every AI agent is surprisingly simple: a while loop, a language model, and some tools.

In this live-coding workshop, we'll build an AI agent from scratch. We'll start from a bare frontier model call to an LLM and incrementally add capabilities. At each step, we'll hit a real limitation, understand why it exists, and solve it. You'll watch a basic chatbot evolve into a sophisticated agent with tools, conversation history, and guardrails.

What you’ll learn

By the end of this workshop, you'll be able to explain how coding agents work and get more out of the ones you use every day.

  • Trace the while loop from a model response to a tool call and back.

  • See how an agent decides whether to continue or stop.

  • Recognize the roles of the language model, tools, and conversation history.

  • Understand where guardrails fit and what limitations they address.

  • Start with a bare model call and add capabilities one at a time.

  • Add a tool and observe how the agent uses it.

Workshop agenda

  • See what makes a coding agent an agent

    Start with one model call. Trace how a harness supplies context, records the conversation, handles tool calls, and decides when to stop.

  • Build the loop with conversation history

    Keep the conversation across model calls and run it to see how each response changes the next request.

  • Give the agent a tool

    Add a file-reading tool, handle the model request to call it, and inspect the result that goes back into the conversation.

  • Inspect behavior and add checks

    Run the agent against a small task. Look at its decisions and discuss checks and limits before letting an agent act on your code.

Learn directly from Matt & Steven

Matt Wynne

Matt Wynne

Cucumber co-founder, co-author of The Cucumber Book, BDD pioneer.

Steven Diamante

Steven Diamante

Helps teams use coding agents with TDD, refactoring, and small steps.

See all products from Matt

Who this workshop is for

  • Software developers who use coding agents and want to understand how they work.

  • All you need is to be comfortable in a modern programming language like TypeScript, Python or Java, and a little curiosity.

Prerequisites

  • Comfortable running TypeScript or JavaScript

    You will edit and run a small Node.js program during the workshop. You do not need prior experience building agents.

What's included

Live sessions

Learn directly from Matt Wynne & Steven Diamante in a real-time, interactive format.

Practice with other developers

Work through the coding exercise alongside other developers, with time to ask Matt and Steven questions.

Maven Guarantee

Your purchase is backed by the Maven Guarantee.

Frequently asked questions

Maven for Teams

Reimbursement

Get your company to pay

Everything L&D needs: email template, receipts, and certificate of completion.

Get reimbursed

Team discount

Learn with your teammates

Save 20%+ when 2 or more teammates enroll in the same cohort.

Save 20%+ with a team

Private cohort

Run a cohort for your org

A dedicated cohort with a custom schedule and curriculum, tailored to your team.

Book a private cohort

$850

USD

Dec 2
·

3–6:30pm EST

Enroll