One engineer, ten years, and a strong preference for things that work.
I’m Patrick Kenneally. I’ve spent a decade building production software — most of it for organisations where the thing failing in public would be a genuine problem rather than an inconvenience.
Most recently that has meant architecting and leading the UK national volunteering platform for Royal Voluntary Service: empty repo to national launch in eight months, then 175,000 concurrent users arriving in a few minutes because our advert ran on ITV. Before that, education marketplaces, real-time trading systems, broadcast infrastructure and a long freelance apprenticeship in small-business web work.
Paddy Systems LTD is the consultancy version of that. It is deliberately small: you talk to the person writing the code, because that is the arrangement that produces good software.
175,000
peak concurrent users
1M+
concurrent load verified
8 months
empty repo to national launch
10 years
shipping production software
Where the experience comes from.
Royal Voluntary Service
Technical Lead & Architect
Architect and technical lead for GoVo.org, the UK's national volunteering platform. Built it from an empty repository to national launch in eight months, and held 175,000 concurrent users through an ITV campaign. Led five engineers.
Learningful
Senior Software Engineer
Rebuilt an underperforming education marketplace into a commercial platform: decoupled architecture, horizontal scaling, and a monetisation model that actually shipped.
Freelance
Distributed & real-time systems
Contract engineering on distributed ledger applications. Real-time arbitrage detection across exchanges, transaction control, rate limiting — in an environment where defects are financially exploitable.
Dotfive
Software Engineer
Business management suite, operational GUIs for UK national smart meter rollout, brand platforms for Toshiba, Samsung and LG, and stream-splitting infrastructure for international sport.
WebWorks
Freelance developer
Full-stack delivery for commercial clients — custom builds, CMS, e-commerce, and revenue-linked SEO and monetisation work. Where the consulting instinct came from.
Four things I will not compromise on.
Boring technology, chosen deliberately
The interesting decision is usually which problem to solve, not which framework to use. I pick tools with large communities and long support horizons, and I write down why in an architecture decision record you can read later.
You should be able to fire me
Infrastructure as code, documentation, no proprietary runtime, no bus factor of one. If the relationship ends, you keep a system your own team can operate. That is the point.
Scale is a design problem
Almost every performance issue I have fixed was structural — a query pattern, a render path, a cache boundary — not a smaller instance size. Load test early against realistic traffic, and the expensive surprises stop being surprises.
Honest estimates over comfortable ones
If I think your idea will take twice as long as you hope, or that you should not build it at all, I will say so on the first call. It costs me work occasionally. It has never cost me a client.