ABOUT
I’m a Senior SWE & AI at Axon. Most of what I’ve done in AI comes down to one problem: a model works in a notebook, and then it has to keep working once real people and real data hit it. A lot of that work isn’t AI at all. It’s the backend, the services and the infrastructure around the model, and that’s usually where things break.
The AI Merge is where I share what I’ve picked up along the way, in guides, diagrams and courses for engineers moving into AI.
Multimodal systems
Models that work with video, images, audio and 3D, and the pipelines that make that data searchable.
Hardware and optimisation
Getting models to run fast enough and cheap enough, on edge devices and in the cloud.
Agentic systems
Agents that act on real operational data, and keeping them predictable once they can change things.
A few things I’ve worked on.
Published computer vision research
Peer-reviewed and published.
Perception for autonomous driving
A perception prototype for a self-driving BMW X5, running on an NVIDIA Jetson AGX.
In-store vision at production scale
Vision systems running in physical retail stores.
AI enablement for engineering teams
Helping teams at work pick up AI tools and coding agents without letting code quality slide.
Freelance ML engineering
Top Rated ML engineer on Upwork, with around 20 projects delivered on and off the platform.
I draw the whole system before I write any code. If I can’t draw it, I don’t get it yet.
Then I go part by part: what are the options, what does each one cost, how does it break. I’d rather show you why than hand you a recipe that only works on my machine.
Agents come in last, one piece at a time, and I read every change they make.
Finishing The AI Atlas, a visual guide to the whole AI stack, and sharing Field Notes on Substack along the way.