Software Developer & Automation Engineer
I build dependable systems that turn noisy real-world input into decisions people can act on.
Open to UK roles and worldwide remote opportunities
I’m a London-based Software Developer who builds practical, user-focused solutions to real-world problems. I care about scoping a problem precisely, proving that what ships actually does what it claims, and building software people can rely on.
Recently I built parts of the OnisAI offer-scoring platform, including its OCR pipeline, contextual scoring engine, backend workflows and dashboard UI. That work led to a second system: a route corridor engine that scores a journey by what it crosses, not only where it ends.
My route into engineering came through product design and operational work. It shaped how I build: make uncertainty visible, design failure paths first, and prefer evidence gathered from real use over rules that merely sound convincing.
May — Jul 2026
Built a field-ready iPhone automation system for low-friction remote execution, including desktop-to-phone controls, self-updating GitHub delivery, diagnostics and recovery tooling.
Oct 2025 — Jun 2026
Delivered OCR ingestion, parsing and validation, a contextual offer-scoring engine, backend reporting workflows and dashboard UI for a multi-tenant driver platform.
May 2023 — Sep 2025
Designed clear digital products and web experiences, translating operational needs into usable interfaces and maintainable design systems.
May 2017 — Feb 2022
Customer-facing operational work in Greater London — experience that still informs how I design tools for people working under real constraints.
2021 — 2023
Combined business operations, management and analytical work — experience that now informs how I judge software by its operational effect, not only its implementation.
A spatial-temporal graph built from driver-observed traffic data. It intersects a live route with exact, near and time-matched corridor evidence so the path matters as much as the destination.
A multi-tenant platform that turns a rideshare screenshot into an explainable accept or decline verdict using resilient OCR, validation fallbacks and context-aware scoring.
Photograph any gym machine and turn it into a workout. GymLens identifies the equipment, shows you what exercises it supports, and carries everything into a tracked session with progress you can see.
Automation pipeline
Multi-stage PDF extraction into validated JSON, reducing manual review by roughly 80%.
Workflow automation
Classifies inbound files, extracts metadata, runs OCR and routes each document onward.
What comes next
I am currently open to software and automation roles across the UK, plus remote opportunities worldwide.