About


About
How I got here
I started writing code at 12, messing around with YaBasic on a PlayStation 2. Sony had shipped the PS2 as a "home computer" to dodge import taxes, but for me it was a gateway into software that's now lasted over two decades.
Since then I've worked across the stack at real-stakes companies — crypto trading at Archax, collateral management at CloudMargin, digital banking at Nationwide, dealer software at CDK Global (seven years), and most recently AI chatbots at OneBot. I've been a software engineer, a technical lead, a Scrum Master, and the person who gets called in when something's broken and nobody can figure out why.
Twenty years in, the thing that's stayed consistent is the same instinct: cut through the noise, find the decision that actually matters, and make it.
What FortyTech is
FortyTech is how I work with startups now. FortyTech offers Fractional CTO engagements — senior technical leadership without the cost or commitment of a full-time hire. The differentiator is AI enablement — alongside the leadership work, I install the AI-assisted workflows that actually make engineering teams faster. Custom Claude agents, slash commands, debugging loops, scaffolding patterns. And I train the team on real usage, not theory.
Engagements are structured around two productised entry points:
- AI Acceleration Sprint — fixed-scope, 2–3 weeks. I embed practical AI workflows into your engineering team: custom agents, integrations into your tracker and observability tools, and patterns that stick once I'm gone.
- Technical Direction Reset — fixed-scope, 2–3 weeks. I go deep on your product, architecture, and team, and give you a clear 30/60/90 plan for what to build next — with the hard decisions made and the rationale documented.
Both fit naturally into an ongoing Fractional CTO retainer if the fit is right. Example shape: weekly leadership call, async access between, ongoing technical decisions, unblocking, and continued AI workflow evolution — adjustable per engagement.
Worth knowing: the harder install isn't the technical one — it's persuading a sceptical engineering team that AI tooling is worth their time. I've taken a team through that conversion directly: leadership-led adoption, deliberate rollout, and the engineers who pushed back hardest are now the daily users. That's the install pattern the Sprint productises.
How I work
My focus is the work that compounds — clarity on what to build, architectural decisions that scale, and the AI workflows that make small teams move faster. Less "extra hands to code," more of:
- Technical direction — what to build, what not to, in what order, with what tradeoffs.
- Architecture ownership — system design, standards, the decisions that compound.
- Delivery acceleration — unblocking, simplifying, AI workflow installation.
- Team guidance — hiring input, levelling, structural shape.
Engagements are remote, UK hours, with selective in-person time when it actually helps. I take on one to two new engagements at a time.
Who I'm a good fit for
The fit is best with:
- Early to mid-stage startups, 2–10 engineers, founder-led technical decisions.
- "Things feel messy" / "we're slower than we should be" / "we've made some questionable tech decisions."
- Want clarity and confident decisions, not just advice they have to process themselves.
Less of a fit:
- Teams looking for raw engineering capacity — that's a different shape of arrangement than what this is.
- Solo founders without an engineering team yet — there's no team to install workflows into; come back when you've hired your first two engineers.
- Engineering teams of 20+ — the Sprint is designed for small teams (2–10 engineers) where the install can reach everyone in the room; larger orgs need a different product shape.
Tech I know well
Frontend: React, Next.js, TypeScript, Tailwind, web3/Ethereum integrations Backend: Node.js, TypeScript, Express, C#, .NET Databases: Postgres, MySQL, Redis, DynamoDB, MongoDB Cloud: Mostly AWS (Lambda, SQS, SNS, ECS, DynamoDB) AI: Claude (Anthropic) for agentic workflows and production integrations, OpenAI APIs, vector databases (Qdrant, Pinecone) Other: WebSockets, real-time systems, API design, testing, observability
I can work with other tech when it's the right call, but these are what I reach for when I'm building or reviewing from scratch.
Elsewhere
I publish at The Random Coder — short videos on AI-assisted coding and contrarian takes on familiar problems.
Get in touch
Use the form on the homepage — name, company, what you need, rough timing. I'll reply within two business days.