AI-augmented
AI-augmented product delivery
Any language, any platform. Principles + AI = shipped.
Home of the thinkers who do, and doers who think.
Engineering excellence × AI leverage. From idea to production.
Thirteen years of software engineering — now applied with AI as the default lever. Any language, any platform, rapid prototypes that become production. Humans in the loop only where it counts: plan approval, MR approval, exploratory testing. AI handles the rest.
AI-augmented engineering. Across languages and platforms.
Principles of programming and software engineering do not change. With AI as leverage, anything can be built — in any stack, faster and to higher quality than the traditional way.
Speed
Prototype to production
Idea to working prototype in hours. Prototype to production through the same pipeline.
Bespoke
Tailored applications, any platform
Android, iOS, macOS, Linux, Windows, web — your own software, fitted to your processes exactly.
Automation
Autonomous delivery pipelines
AI writes the plan, code, tests, and review. Humans approve the plan and MR. That is it.
Cloud
Enterprise cloud systems
AWS, Azure, GCP. Multi-region, multi-tenant, compliance-ready.
Autonomy
Autonomous AI agentic teams
Multi-agent systems. Specialised agents, one goal.
Lean pipeline. Maximum delivery, minimum ceremony.
AI does the work. Humans set direction, approve plans and check outputs. From idea to production — in eight steps.
- 01 Human
Idea
You bring the problem or vision. A few sentences or a sketch is enough — context fills in at the next step.
- 02 AI
Analysis
AI ingests existing product documentation, the codebase, user flows and design language. It maps context and constraints before proposing anything.
- 03 AI
Implementation plan
A comprehensive plan grows out of that analysis: what changes where, in what order, with what risks and alternatives.
- 04 Human
Plan approval
You review, adjust or reject the plan. No code is written without an approved plan.
- 05 AI
Development
AI implements the approved plan — code, refactoring, documentation. All in an isolated environment with clear guardrails.
- 06 AI
Testing
AI runs unit, integration and E2E suites. Generates new tests as the change demands and iterates until they pass.
- 07 Human
Review
You inspect the diff, walk through the merge request, do exploratory testing where intuition matters. Approve or send back for changes.
- 08 Human + AI
Deployment to PROD
AI executes the deployment with automated observability and instant rollback. You sign off on the release window and watch the first minutes in production.
→ Result: a small team ships like a large one. Throughput decouples from headcount.
Built in real production.
Companies where I shipped and led teams. Betting, payments, financial services, B2B platforms.
-
Sky 2013–2015 -
Sky Betting & Gaming 2015 -
Infinity Works 2015–2016 -
NHS 2016–2017 via Infinity Works -
T. Rowe Price 2018 via Infinity Works -
Capco 2019–2023 -
Paysafe 2019–2023 via Capco -
group.one 2023–present
Three principles.
01
Principles over stack
Languages and frameworks change. Engineering principles do not. Once they are in place and you use AI as leverage, picking a stack becomes an implementation detail. Rust today, something else tomorrow — the principles stay.
02
AI does the work, humans gate decisions
Implementation plans, merge requests, exploratory testing — these need human judgment and context. Everything else — code, tests, documentation, first-pass review — AI handles faster, and often better.
03
Prototype and production share one pipeline
Rapid prototyping is not a separate process from production. The same AI-augmented workflow ships a prototype in hours and pushes it to production without a rewrite round. Speed with no quality trade-off.
Have a concrete problem or vision?
The best projects start with a half-hour call. Book a slot — I will look at your context and tell you straight whether I can help.