AI services
AI built for a specific job in your business, and maintained after launch.
We tell you first whether AI is the right tool, because often a better report solves it for nothing per use. When it is right, it runs against your own documents and records, on infrastructure we manage, at a cost measured per request before you commit.
Overview
AI, the way we do it
Much of what is sold as AI is a demonstration. The question that matters is whether a model still answers correctly next month, on your data, at a price you can predict. So we start by assessing whether AI is the right tool at all. Often a well-written report solves the problem faster and costs nothing to run.
When it is the right tool, we build it against your own information: your documents, your records, your systems. It runs on infrastructure we configure and watch ourselves, so nothing leaves your control that you did not agree to send. This is the newest of the seven things on this list for us. The data and server work underneath it we have done for years, but none of the case studies on this site is an AI project yet.
Under this service
What AI covers
- Technology AI AI that does one job, properly.
- Solution Machine Learning Models that earn their place in the business.
- Solution Data Science Answers from the data you already have.
- Technology LLMs Language models, kept on a short leash.
- Technology Generative AI Generation with a human still holding the pen.
- Technology Python Python, for the work that has to be read as well as run.
What's included
What you actually get
- A clear answer first on whether AI is the right tool
- Assistants that answer from your documents, not the open internet
- Paperwork read and sorted automatically, without manual re-entry
- Connected to the systems you already run
- Cost measured per request, so the monthly bill is predictable
AI capabilities
How we work
Five steps, no surprises
- 01
We talk
A call or a WhatsApp thread. You tell us what is not working; we tell you honestly whether we are the right people.
- 02
We scope it
A written plan with what you get, what it costs and how long it takes. Fixed, so there are no surprises later.
- 03
We build it
You see it as it goes, not at the end. Changes are cheap while it is still being built.
- 04
We put it live
On infrastructure we set up and secure, tested before the launch date.
- 05
We keep it running
Updates, monitoring and someone who answers. Most clients stay on a monthly agreement.
Questions
About ai
Do we actually need AI for this?
Often, no, and we will say so. A search box, a better report or a fixed set of rules solves many of the problems AI is suggested for, at no cost per use. We would rather establish that at the start.
Where does our data go?
Wherever you agree it goes, and nowhere else. Models can run on a server we manage for you, or through a provider under a contract that forbids training on your content. If the data has to stay inside Saudi Arabia, say so at the start and we will build for that.
What stops it making things up?
Grounding it in your own material and having it cite its source, so any answer can be checked. And a limit on what it is allowed to answer at all: a system that declines to answer is more useful than one that answers incorrectly.
What does it cost to run every month?
It depends on volume, and we measure it per request before you commit rather than after. You get a projected monthly figure with a ceiling, and monitoring that tells us if real usage starts drifting above it.
Tell us what you have in mind.
An engineer reads your message and replies to you directly.
Talk to us