Services
Three kinds of engagement — one way of working
We scope every project around your existing tools and actual workflows. We do not build towards a theoretical best state that requires a six-month IT migration to reach.
01 — Project
Agent Design & Build
For teams with a specific automation target: a repetitive process that costs hours every week, a decision that could be made by a well-designed rule set, or a workflow that currently requires three tools and a human in the middle.
We start with your environment — what tools you have, how data moves between them, where the actual friction is. The agent we build is a response to that specific situation, not a template applied to it.
Projects typically run 6–10 weeks. We commit to a fixed scope and price after the initial scoping call. We do not do open-ended project work.
Typically includes
- Discovery session to map the current workflow
- Agent architecture proposal with rationale
- Integration design for your existing tools
- Build, testing, and edge case handling
- Documentation for your team
- 30 days of post-launch support
Starting from €3,500. Final price depends on integration complexity. We quote after the scoping call.
02 — Retainer
Ongoing Agent Maintenance
Agents are not static. Tools update their APIs. Your internal processes change. Model behavior shifts in ways that are not obvious until something downstream breaks. Most consulting relationships end at delivery. This one does not.
Our retainer clients have a direct line — a real person, not a support ticket queue. When something breaks, we are already aware of the context because we built it.
Available to clients whose agents we built, or to teams inheriting AI systems they did not design — after a technical onboarding session so we understand what we are maintaining.
Typically includes
- Monthly monitoring and health checks
- API and dependency updates as needed
- Incident response within one business day
- Quarterly review call to assess drift and improvements
- Priority access for new automation requests
From €1,200 per month. Minimum three-month commitment. Cancel with 30 days notice.
03 — Integration
LLM Integration into Existing Products
You have a working product and want to add language model capabilities to it. A document assistant. A search feature that understands context. A chatbot that knows your product's specific domain.
This is trickier than it looks. LLMs behave inconsistently at the edges, and the edges are where your users live. We scope carefully and we are honest about where the technology is reliable and where it is not.
We work in TypeScript/Node.js and Python. We have integrated with OpenAI, Anthropic, Mistral, and open-source models running on-premise.
Typically includes
- Use case assessment — where LLMs help and where they don't
- Model selection and prompt engineering
- API integration into your existing codebase
- Evaluation framework to track output quality over time
- EU data residency options where required
Scoped per project. Starts with a technical discovery session. Quote provided after.
Not sure which applies to you?
Start with the contact form. Describe the situation in plain language. We will tell you how we would approach it and whether it is something we take on.
Reach us