What we do

Five disciplines, run by one team. Most projects use more than one of them, and the work is not handed between departments on its way through.

01
Engineering
Applications, APIs and the infrastructure they run on.
This is the centre of the work. Services with clear boundaries, interfaces agreed before anything depends on them, and infrastructure written down as code so an environment can be rebuilt rather than remembered. We build for the traffic a system will actually see, and for the person who has to change it in two years.
What you get
- Web and mobile applications, and the services behind them
- APIs, integrations and infrastructure set up as code
- Moving an existing system across, without a day on which everything changes at once
02
Applied AI
Models and AI features built into software people use.
Applied AI is something we build with rather than a label on the front of the box. The hard part is rarely the model: it is where the data comes from, what the software does when the answer is wrong, who sees a result before anyone acts on it, and how you find out a month later that quality has slipped. We build that part first.
What you get
- AI features inside a product — search, drafting, classification, summaries
- A model taken from a trial to something monitored and running
- A review path for the decisions that should have a person in them
03
Product
Interfaces designed around the task someone has to finish.
Software that is calm to use is a decision rather than luck. Screens are drawn around the task someone is trying to finish, shown to the people who will use them every day, and cut back until what is left is obvious. The measure is how little training the result needs.
What you get
- Interface and interaction design
- Prototypes put in front of the people who will use them
- A design system your own team can build on without asking us
04
Data
Reporting and pipelines that answer a question you ask.
A business produces evidence all day and throws most of it away. We build the path that keeps it: imports that fail loudly instead of dropping records quietly, one agreed definition for each figure, and reporting on top that answers a question somebody actually asks.
What you get
- Imports, pipelines and one agreed definition per figure
- Reporting and dashboards fitted to how you work
- Forecasting, where there is enough history behind it to mean something
05
Operations
Hosting, monitoring and looking after what we deliver.
Most software gets into trouble after launch rather than during it, because nobody owns it. Looking after what we build is part of what we do: hosting, updates, monitoring that answers questions instead of raising them, and a route to somebody who knows the system.
What you get
- Hosting, monitoring and updates under an agreed level of cover
- Fixes and small changes once a system is live
- Handover — runbooks, notes and training — whenever you want to take it in-house

Contact

Tell us what you need built.

A website, an app, an online shop, an internal system, or a change to something you already have. Write to us and a person answers.

hello@rysellc.com Book a call

What we do

Engineering

Applied AI

Product

Data

Operations

Tell us what you need built.