Most of my projects begin with a question.
Sometimes it starts with a problem that keeps getting in the way. Sometimes it is a technology I want to understand properly. Occasionally it is a slightly unreasonable idea that I cannot leave alone until I have built enough of it to find out whether it works.
This is where I document those experiments: what prompted them, what I tried, what failed, what changed my mind and what I learned along the way.
These projects span product engineering, AI, knowledge graphs, games, systems programming and a few ideas that probably should have remained safely theoretical.
They are connected less by technology than by the way they started: I noticed something, wondered whether it could work differently and built enough to find out.
I wondered if AI was designed into the development lifecycle from the beginning. What would engineering delivery look like?
An attempt to build delivery tooling with AI in the workflow from the start — readiness scoring, role-based review agents and human-approved write-backs — with Azure DevOps kept as the system of record rather than replaced.
2024 – Present
I wondered if history could be explored as a connected temporal system rather than a collection of separate articles?
A project that began with a simple request for a more joined-up way to explore history and grew into a temporal knowledge graph and mobile application.
2025 – Present
I wondered if can a small set of mechanics produce a mobile rogue-like game that is simple, fair and difficult to put down?
A mobile game experiment focused on pacing, replayability and the uncomfortable discovery that balancing a simple game is not simple at all.
2026 – Present
I wondered if we redesigned the web today using everything we've learned since it began, What would the web look like?
A long-running thought experiment exploring what a modern web could look like if identity, trust, provenance and user ownership were designed into the network from the beginning.
2019 – Present
I wondered if you simulate a field of coupled oscillators, can you make the patterns in it mean something?
An attempt to build a computer out of a wave simulation, then teach it to reason, then let a language model write programs for it. It ran, it taught me a great deal, and it stopped after ten weeks.
2025 – Present
I wondered if the board game I loved as a child be rebuilt as a 3D digital tabletop without turning into a video game?
An attempt to recreate the original 1989 HeroQuest as a faithful 3D tabletop game in Unity, with the rules encoded as a testable engine, the board sitting on a virtual table, and the miniatures generated by AI. Playable and unfinished, in that order.
2025 – Present
I wondered if intelligence didn't live inside one giant model, but emerged from a persistent network of smaller specialised modules?
An experiment exploring whether a modular cognitive architecture can learn continuously, adapt its own structure and retain experience without relying on a single monolithic model.
2026 – Present
I am usually at my best when the answer is not obvious yet.
I like taking a loose problem, turning it into something testable and building the first working version. That might mean a technical spike, a reference implementation, a reusable platform component or a small tool that removes friction for everyone else.
I have led engineering teams, but I have always remained hands-on. I enjoy moving between product questions, architecture, application code, cloud infrastructure and whatever unfamiliar area the problem happens to require.
I learn by building, and I build because I am curious.
A changing list of things I am pulling apart, joining together or attempting to understand properly.
Some of these projects have public repositories. Others are private, incomplete or contain work that needs a guided explanation to be useful.
Where I can share source code, I will. Where I cannot, I will show the decisions, diagrams, prototypes and lessons that explain the engineering behind it.