Over the past year, I've shipped in five AWS hackathons and challenges. Kiro CLI let me move faster than I ever had — but vibe coding still didn't work. Sessions would spiral into hallucination loops I couldn't trace back or recover from. That's what pushed me to micromanage Kiro CLI instead of vibing with it, starting with the AWS Lambda Hackathon a year ago and every initiative since. Working this way not only let me ship a lot more — I've never had to look back.
For the first time in my career, the team I was working with was able to provide a reliable implementation of a feature at the same time the concept was being discussed. Not just one feature, but a collection of them that interacted with each other as a cohesive product. After six hours working with a team that had met for the first time, we had a working alerts map for Wellington City. That only works if the process can absorb changing your mind mid-build without losing the thread.
Humans need clear visuals and AI needs a clear format. I couldn't find an Obsidian plugin that provided both and was actively maintained, so I built my own.
An always-on agent that turns a scanned piece of sheet music into a sequenced, lesson-by-lesson practice plan, triggered entirely by moving a card on a kanban board.
Talk to your markdown todos. An Alexa skill powered by Lambda, S3, and Amazon Bedrock — built for the Weekend Productivity Challenge.
AI assistants today collaborate on their terms. Lucid Thread gives you twenty specialized AI personas — or build your own. Visible memory. Real-time document tools. All operating within boundaries you design, not boundaries designed for you.
Three years ago I got curious about AI. Not the headlines — the actual thing. What it could do. How it worked.
So I started entering hackathons. Five so far, two more right now.
And somewhere along the way I realized the most valuable thing I built wasn’t any of the projects.
It was something I almost didn’t notice.
AI has lowered the barrier to execution. It’s easier than ever to prototype, automate, and ship. In many cases, you can get surprisingly far just by prompting well. That’s powerful. But I’ve been thinking about something else.
This is the companion to The Case for Change, which covers the why — the urgency of the exponential economy, the psychology of resistance, and the principles behind successful adoption. This article is the how.
Knowing your organisation needs to adopt AI and actually making it happen are two very different problems. The technology is almost never the hard part. Habits are.