The AI Operating System for your team.
AI isn’t a tool you install. It’s the layer your business runs on, once it’s in properly. I help teams figure out what belongs in that layer, build the bits that matter, and train the people who’ll keep it running after I’ve gone.
Want to DIY? Start with our articles. If you’re not quite sure what you’re looking for, our consulting page is the place to begin. If you’d like to talk, get in touch.
Get Started
Bespoke AI R&D, inside your business
A few months of forward-deployed engineering. I watch how the work really flows, figure out what belongs in your AI layer, and build the systems your team runs after I’ve gone. No two engagements have looked the same.
How I work →Things I’ve built. Systems I’m running.
Open-source MCP servers, SaaS products, and websites. Some power my own businesses. Some are used by other developers. All of them are running today, in production. If any of it is useful to you, or sparks an idea for something you’d like to build, I’d love a chat.
See the work →Write-ups from the work
How-to guides, notes from client engagements, opinions on where AI is heading. No black-box secrets, because I think AI is for anybody willing to learn. Most of these come from problems I’ve hit on real work.
Read the articles →Recent writing
-
Agentic Interoperability for Website Owners: AI User Experience (AI UX)
I’ve been keeping a close eye on the emerging subject of “agentic interoperability” across all my recent build projects. In layman’s terms, that’s the ability for AI agents to do things for you on your behalf – particularly inside SaaS apps via MCP servers, and increasingly via WebMCP, an emerging protocol that fills the gap
-
Implementing WebMCP on a Recruitment Website
Thinking about what, exactly, the future of a website “looks” like in the agentic era is a challenging proposition. It might be that in most cases, our future viewers/readers/customers can do everything, from their chatbot of preference while never visiting your site. WebMCP is a part of this puzzle. The protocol directs an agent to
-
How to Set Up a Claude Code Project (And What Goes Where)
The .claude folder is the control centre for how Claude Code behaves in your project. Here’s what goes in it, what each file does, and the step-by-step setup I use for every new project.
-
Swapping the Engine: How to Run Claude Code on Local Silicon for Zero Pennies
Claude Code’s real power isn’t the Anthropic model sitting behind it, it’s the agentic : the file-system access, the tool use, the way it chains tasks together without you babysitting every step. I figured this out the expensive way. I ran a batch of log-parsing scripts through the API for a client project last month
-
Claude Desktop System Requirements: Windows & macOS
Have you found yourself becoming a heavy AI user? For Claude Desktop, what hardware matters, what doesn’t, and where do Anthropic’s official specs look a bit optimistic? In this article: Official Requirements | Windows vs macOS | What Actually Matters | RAM | MCP Servers | Minimum vs Comfortable | Mistakes Official Requirements Anthropic doesn’t
-
Best GPUs for Running Local LLMs: Buyer’s Guide 2026
I’ve been running various LLMs on my own hardware for a while now and, without fail, the question I see asked the most (especially on Reddit) is “what GPU should I buy?” The rules for buying a GPU for AI are nothing like the rules for buying one for gaming – CUDA cores barely matter,
Want a chat?
45 minutes. Tell me what you’re trying to figure out. I’ll be honest about whether I’m the right person or whether you’d be better served somewhere else. No pitch deck, no follow-up sequence.
Get in touch