01
Software engineering
Years of building software that businesses depend on every day. Architecture, APIs and integrations, testing, and the less glamorous work of keeping systems reliable once they’re live.
Founder · AI engineer · Hull
Liam Silk.The engineer you’ll work with.
I’m the founder of Nirox Labs, based in Hull, and the engineer who’ll design and build your system. I started Nirox Labs to build custom AI for businesses: systems that work with your data, fit the way you already work and still make sense after they’re handed over.
10+ years’ experienceYou speak to the person building itBased in Hull, working across the UK
What I bring
Engineering depth,
explained plainly.
02
Businesses of every shape
I’ve built software for businesses of all kinds and sizes. Different sectors, same lesson: good software starts with how the work really happens.
03
AI engineering and solution design
Designing systems around language models: retrieval over company knowledge, agents with permissions and approval points, evaluation against real examples, and the architecture that connects them to the tools you already use.
04
Training and adoption
I have a background in training people and helping teams adopt new tools and ways of working. A system only delivers value when people understand it and trust it, so that’s built into every project.
Why training matters
Built to be used,
not just delivered.
Plenty of technology projects work on the day they’re delivered and quietly fall out of use a few months later. Usually the problem isn’t the software. It’s that the people using it were never properly brought along.
My experience training teams shapes how every Nirox Labs project runs. Your team is involved while the system is tested on real work, so they see how it behaves and where it needs their judgement. Handover includes documentation written for the people using the system, not just for developers, and training sessions for your team.
The aim is simple: when I hand a system over, your team knows how to use it, when to trust it and what to do when something looks wrong.
I use the same approach on my own business. Nirox Labs runs on an AI operating system I designed myself, connecting business context, a record of every interaction and reusable skills, so proposals, research and follow-ups draw on everything that came before. Read how it was built.
How I work
A few principles
behind every build.
Start with the work
Understand the process, the people and the data before choosing a technology. Sometimes the answer isn’t AI at all.
Small and useful first
A focused first version that proves its value on real examples beats an ambitious plan that never ships.
Built to be checked
Sources shown, actions logged, approvals where they matter. You should always be able to see what the system did and why.
Yours when it’s done
Documented, handed over and understandable to your team, so it doesn’t depend on me to keep running.
Tools I use
Chosen for the job,
not the trend.
A selection of the languages, platforms and services I work with. Each project uses whatever fits your systems, data and running costs best.
- TypeScript
- Node.js & Express
- Vue & Svelte
- REST APIs & integrations
- PostgreSQL & Supabase
- Trigger.dev
- Claude & the Anthropic API
- Claude Code
- n8n
- Unit & evaluation testing
A useful first conversation
Let’s talk about what you want to build.
Bring the problem, the idea or the process. You’ll speak to me directly, and I’ll tell you honestly whether it’s worth building and what a sensible first step would be.
Discuss your AI project Read the free RAG guide Or email hello@niroxai.comIndependent AI engineering. Based in Hull & East Yorkshire, working across Yorkshire and the UK.