Applied AI Product Engineer and hands-on engineering lead with 13+ years shipping production systems across fintech, digital assets, and AI products — built with agentic workflows, RAG, and LLM integrations.
I design and ship applied-AI products end to end — from the first prototype to the production system real customers depend on. Across 13+ years I've built for fintech, digital assets, enterprise platforms and, now, AI-native products, with a strong TypeScript, React, Next.js and Node.js foundation underneath.
Today that means agentic workflows, RAG and LLM integrations — built with LangGraph, LlamaIndex, the Vercel AI SDK and vector databases — wired into interfaces that stay fast, accessible and reliable under load. Five of those years were spent in London; now I work from Abu Dhabi.
“Turning ambiguous product problems into reliable, user-facing software.”
Three overlapping disciplines. Each one is about the same thing: getting capable software safely into the hands of real users.
Taking AI from prototype to production — agentic workflows, retrieval-augmented generation and LLM features built to be observable, evaluated and reliable rather than demo-ware.
Turning ambiguous product problems into reliable, user-facing software — shipping AI-powered features end to end on a strong TypeScript, React, Next.js and Node foundation.
Designing robust, accessible, performant frontends — monorepos, Core Web Vitals, testing and CI/CD pipelines that keep large systems scalable and maintainable.
A working model of how I run multi-agent delivery — Claude Code, Cursor, LangGraph and OpenCode in parallel sandboxes, moving through request → routing → execution → review → merge.
Parse intent, load project context
Select agents via the state graph
Parallel work in isolated sandboxes
Automated review, tests, validation
Squash and integrate the branches
Checks green — ready to ship
A pragmatic, deep toolset rather than a buzzword list. These are the things I reach for first, and the supporting cast around them.
Four working principles that hold whether I'm wiring an agent graph or sweating a hundred-millisecond interaction. They are the difference between software that demos and software that ships.
A demo proves it can work; production proves it does. The interesting engineering lives in the gap between them — evals, fallbacks, latency and the unglamorous edges.
Systems are built from small, composable primitives. Decoupled pieces stay legible, testable and replaceable — and keep a codebase from calcifying into technical debt.
LLMs are treated as load-bearing, not novelty. Workflows are designed around autonomous agents with deterministic guardrails and graceful degradation when models surprise you.
You can’t improve it. Telemetry, tracing and evaluation are part of the foundation — observability is designed in from the first commit, not bolted on after launch.
Roles across fintech, digital assets, enterprise platforms and AI products. Five of them in the UK; the most recent firmly inside applied AI.
Drive applied AI product development end to end — designing, prototyping, and shipping AI-powered solutions that connect real customer needs to production-grade engineering, with measurable adoption and business impact.
Drove product innovation at cMatter, an AI and advanced-ML platform for building energy management, from idea to launch.
Led frontend architecture and delivery for a regulated fintech investment and trading platform built in React and TypeScript.
Built and maintained Copper's institutional digital-assets and custody platform, integrating with complex backend and blockchain systems.
Built secure, robust, responsive web applications and mentored within an agile team to ensure high-quality delivery.
Engineered large-scale, transformative digital solutions for enterprise clients.
Delivered digital experiences for clients and bridged design and engineering teams to protect UX and requirements through delivery.
Built secure web products and UI for startups; owned application architecture and UI implementation.
Built websites, WordPress themes, and web applications for startup clients; conducted market research and owned UI implementation.
I'm open to select consulting engagements, applied-AI architecture work and high-impact engineering roles. If you're taking AI from prototype to production, I'd like to hear about it.