Build a Software Factory: Hands-off agentic coding for experienced engineers
Matt Wynne
Cucumber co-founder, BDD pioneer
Aldric Giacomoni
ex-General Assembly, XP agentic builder
From prompting the code to engineering the factory that generates it
Use promo code early-bird-jan-26 for $500 off until 30th November.
It's easy to get agents to generate code. But afterwards you're left with a backlog of PRs of varying quality that still need your full attention. Does this mean you're destined for a future as "Code Reviewer in Chief"?
Imagine a world where your agents consistently generate trustworthy output that you can confidently deploy without reviewing every line.
We've lived in that world and we want to take you there!
This course is for you if you're:
Spending as much time correcting the agents output as you would have spent writing it yourself
Working in a brownfield codebase with landmines all around
Drowning in agentic PRs
Already running agents in a loop, and spending your evenings babysitting them
Over four weeks you'll build your own software factory:
Shaping work into iterations the factory can finish
Keeping the machine that writes the code away from reviewing it
Seeing what a run is doing without reading the code
Putting a foreman between you and the factory instead of babysitting it
Simplifying a factory that has stopped converging
Building a factory takes real time and tuning. Go deep with experts who hav been there.
What you’ll learn
Go from reviewing everything your agents write to engineering the factory that writes it, and trusting what comes out.
Spend more time up front on requirements, and write them as skills the factory can follow rather than prompts you retype.
Break a project into iterations small enough for an assembly line to complete.
Spot the iteration that looks reasonable to you but is still too big for the machine.
Build a separate validator into the assembly line, because the machine that just wrote the code is the wrong one to review it.
Add a different model as a critic to widen coverage, since agents on the same model share blind spots.
Close the gap with an external check, such as a test or a schema, because agreement between models is not verification.
Build the views that show how a run is going, the way a thermostat or an oscilloscope would.
Connect the factory to your real systems with MCP, so those views draw on what is actually happening.
Decide from those views when to step in and when to leave the run alone.
Put a foreman between you and the factory, an agent that helps you shape the work, start the line and read what changed.
Design where the agent acts, checks, escalates, or hands control back to you.
Manage each machine's context window deliberately, so a long run does not drift as it goes.
Tune and calibrate the factory, and take parts out when a run stops converging. A more complicated factory is not a better one.
Treat the factory as the thing you engineer. Your attention moves from the code to the system that builds it.
Learn directly from expert instructors
Matt Wynne
BDD pioneer, Cucumber co-founder, and creator of Example Mapping.
Aldric Giacomoni
Agent-native builder and educator shipping production software with AI agents.
David Laing
Bringing agents to SBOMs, vulnerability triage, and remediation.
Jeremy Lightsmith
Executive coach helping leaders navigate complexity in the AI era.
Who this course is for
Experienced, quality-conscious developers who already use AI but don't trust the output without reviewing every line.
Engineers who have already built an agent loop and now spend more time babysitting it than they would like.
Developers who want to engineer the system that writes the code, as well as write code themselves.
What's included
Only in live cohorts
Live sessions
Learn directly from your instructors in a real-time, interactive format.
Lifetime access
Go back to course content and recordings whenever you need to.
Community of peers
Stay accountable and share insights with like-minded professionals.
Live and self-paced cohorts include
Certificate of completion
Share your new skills with your employer or on LinkedIn.
Self paced AI guided tutorials
All theory is presented via guided conversation & exercises you have with your agent so you can absorb content at your own pace prior to discussing it during the live sessions
Hands-on, ~70% building
Most of your time is spent building on your own workflows, not watching slides.
Maven Guarantee
Your purchase is backed by the Maven Guarantee.
Course syllabus
7 lessons • 4 projects
Week 1
Welcome [W1.M01]
Loops [W1.M02]
Week 2
Validation
Assembly lines
Free resources
Testimonials
Matt, Aldric, David and Jeremy share their cutting edge techniques for producing high-quality, rigorously tested and validated software, all using a dark factory backed by the Big Three. They are knowledgeable, industry vets with answers to every question and thoughtful perspective on how to leverage this tech to build software in this modern era.
5/5 rating for this course
Ryan Spore
Staff Engineer
Why this course is different
Frequently asked questions
Maven for Teams
Reimbursement
Get your company to pay
Everything L&D needs: email template, receipts, and certificate of completion.
Get reimbursedPrivate cohort
Run a cohort for your org
A dedicated cohort with a custom schedule and curriculum, tailored to your team.
Book a private cohort$2,450
USD