
The gap is growing fast
We've noticed a fast-growing interest among progressive companies in getting their developers to benefit from agentic coding, done well. There's more and more talk about a gap opening up fast between teams that have embraced agentic development and those that haven't.
The questions we hear most often are:
How do we get people to actually enjoy the agentic style, and get real value from it?
How do we avoid the traps everyone seems to fall into when they start?
How do we make the new way of working stick?
A few ways we can help
Presentation: agentic development inspiration
Q&A with Nizar about Probity and agentic codingNot sure yet whether agentic coding is for your team? Invite Nizar Selander for an hour!
During this presentation he'll walk you through his own journey – from before agentic coding was a thing, to the workflows he relies on today in real, business-critical projects – including a key section of a demo of an AI agent solving a kata largely on its own, kept honest by guardrails from Probity, his own open-source project.
Workshop: two half-days, hands-on from the start
Left alone, agents will happily call something finished without proving it works, or cut corners the moment testing gets in the way. In this workshop we practise TDD and other helpful, proven workflows to keep them in check while amplifying all the good.
We run this as two half-day workshops:
Half-day one: a kata in ensemble format (the whole group programming together), using an AI coding agent inside a TDD loop. You'll see, live, where an agent wants to cut corners and how tests-first discipline stops it. And practice several different ways of working together.
Half-day two: small teams bring a real, small task from their own codebase – a bug fix, a small feature, a targeted refactoring – and drive an agent through it under the same disciplines. Working with real code, on real edge cases, without a kata to hide behind.
Adoption: a journey over a few months
A single workshop teaches the discipline, but making a new way of working stick takes continuous practice and friction removal. Over a few months, we run two tracks in parallel:
Bi-weekly lunch-and-learns, open to the wider organisation: 30–40 minutes on whatever's most useful at that point in the team's journey – a workflow worth spreading, a question that's come up, a pattern worth more people picking up.
Hands-on workshop sessions for a smaller group: we sit with the team closest to the friction and fix it together. Where it makes sense, we turn the fix into a guardrail the team keeps using after we're gone.
Who this is for
This is perfect for software teams and organisations that want to bring applied AI and harness engineering to business-critical work without compromising on long term changeability. It will ensure that your AI-assisted coding is meeting the same engineering standards as everything else you ship.
Let’s talk!
If you'd like to inspire your team with bleeding-edge agentic coding at the start of autumn, we can help. It starts with a conversation about where your team is today and what would actually move the needle.




