Skip to main content

LUNO

GET IN TOUCH
#BACK TO SERVICES

AI

AI ENVIRONMENT SETUP

  • 15+ setups
  • from 1 day
  • tools of our own

We set AI up around your team: access to the models, connections to your internal data and services, ready-made scenarios and instructions. So people reach for it every day rather than once a month.

WHO IS IT FOR?

Three situations where the setup pays for itself in the first week

  • .01

    THE SUBSCRIPTION IS THERE, THE SETUP IS NOT

    Paid for, opened, sitting in the bookmarks. Working with AI comes down to copying code into a chat and back again: the model does not know the project, and the context has to be explained from scratch every single time. The model is not the problem. Out of the box the tool knows neither your code, nor your rules, nor what you have already written — and until it does, it stays an advanced search box.

  • .02

    EVERYONE WORKS THEIR OWN WAY

    Everyone on the team uses AI, but each in their own way: one has their own prompts, another their own subscription, a third a different model entirely. The result depends on who happened to pick the task up today. That is where “AI produces rubbish” comes from — almost always a complaint about the absence of shared rules rather than about the model. Consistent output appears where a team has one configured environment instead of five personal ones.

  • .03

    YOU HAVE HIT THE EDGE OF THE BOX

    The team already works with AI confidently and has run into a ceiling: the model cannot see the internal systems, does not know the domain, and cannot do the one thing you specifically need. From here on it is bespoke — extensions for your stack, access to internal services, scenarios for the tasks you actually repeat. Nothing ready-made exists, because by definition it would have to be about your company.

WHAT'S INCLUDED

We do not build below this line

  • - A review of the project and the stack.01
  • - The environment configured around your code.02
  • - Our own extensions and scenarios.03
  • - Connections to your services.04
  • - Working rules and guardrails.05
  • - Instructions written for your stack.06
  • - A live first run alongside you.07
  • - Support through the first weeks.08
  • THIS IS THE FLOOR, NOT A PACKAGE

    Everything listed ships in every project — even the simplest and cheapest one. There are no stripped-down versions.

  • THE SETUP STAYS WITH YOU

    The configuration, the instructions and the extensions are handed over with the work. Subscriptions are registered in your name, and we keep none of your credentials.

  • ENVIRONMENTS AGE, AND THAT IS FINE

    Models and tools change every few months. We show you what to update yourself and what is worth rebuilding six months from now.

WHAT'S INSIDE

  • THE DATA STAYS WITH YOU

    The model can be deployed on a dedicated server — yours or ours: requests, documents and conversations never reach an outside provider. The infrastructure is run by someone with twenty years behind them, and it is the same infrastructure the rest of our projects live on. If nothing sensitive is involved we work through a provider, which is cheaper and quicker. Which of the two it is gets decided before the start, not discovered after launch.

    THE DATA STAYS WITH YOU
  • AI IS NOT NEEDED EVERYWHERE

    Where accuracy has to hold without exception — calculations, money, legally binding decisions — ordinary code is more reliable and cheaper. A model works well where something has to be read, understood, compared or drafted. During discovery we say plainly which steps stay deterministic and which go to the model. Selling you AI where a script would do is a bad deal for us too: projects like that do not work, and we are the ones who end up explaining why.

    AI IS NOT NEEDED EVERYWHERE
  • EVERY REQUEST COSTS MONEY

    Building it and running it are two separate bills. The second depends on how many users you have and how large their requests are, and it does not go away once the project is handed over. We work out the cost per request before development and set it against the price you charge, across several growth scenarios. And we build in what brings it down: caching repeats, splitting simple and hard requests between different models, limits and alerts when spending spikes.

    EVERY REQUEST COSTS MONEY
  • THE MODEL CHANGES, THE PRODUCT STAYS

    Talking to the model lives in a layer of its own: changing provider, or moving to a local model, does not rewrite the product. This is not future-proofing for its own sake but insurance — over the past two years prices, access terms and the quality leaders have all changed. Plus rules and limits on answers, a journal of every request, and statistics: what people ask, where the model gets it wrong, and what is worth caching.

    THE MODEL CHANGES, THE PRODUCT STAYS

CATEGORIES

Two options. Yours is the one you recognise yourself in

FAQ

Answers to the most popular questions

From $500 for a single developer to $3.000 for a team. The figure depends on how many people there are, how complex the stack is, and how many internal systems it has to connect to. We name the price after a short conversation about the project — here that takes ten minutes, not a separate stage.

A setup for one developer takes a day or two. A team takes one or two weeks, mostly because of clearing access and walking people through it. This is not a project measured in months: you see the result in the very first working cycle, not at the end.

To configure it around the project, yes, in one form or another: without understanding how the code is laid out, the setup is done blind. There are options — repository access, working on your side over a shared screen, configuring against an isolated copy. You choose the format and we agree it before we start. Access is revoked when the work is done, and we keep none of it.

Practically any — the environment is configured around your project, not the other way round. Most of our work is in JavaScript and TypeScript, Python and PHP, but the principle does not change: the setup follows how your repository and your processes are built, not the language. If the stack is an unusual one, we say so straight away.

It will — models and tools move every few months. Which is why we do not just configure it but show you how it is put together: what you change yourself, what can stay as it is, and what is worth rebuilding. A rebuild six months on is a few hours of work, not starting over. Promising that a setup will hold good forever would not be true.

Model subscriptions are registered in your name and paid by you directly — we do not resell access and we take no cut. On the call we will tell you which plan is enough for your volume: overpaying for a bigger one is the most common mistake at the start.

That is a normal situation, and nobody has to be pushed. A configured environment does not cancel the way people already work — it adds a second one. Two or three people picking it up is usually enough: the rest join when they see the difference on real tasks rather than in a presentation.

A course explains how this works in general. We configure what works specifically for you: your code, your services, your tasks. After a course you still have to apply what you learned; after a setup the environment is already standing and running. They are different things, and neither replaces the other.

Every configuration, extension, scenario and instruction. No subscription to us, no lock-in: if tomorrow you decide to work with another contractor or on your own, all of it stays with you in working order. The instructions are written so that someone who was not part of the setup can follow them.