AI integration into existing systems
You don't need to replace your software to use AI. I add features such as summaries, classification or text generation to the systems your team already uses, so daily work continues without disruption.
- Connecting AI models to your applications through clean APIs
- Working with your current data and processes, with no rewrite
- Token costs and data security planned from the start
Private AI and knowledge search
Company knowledge often sits in documents nobody has time to search. A search system built on your own files lets your team ask questions in plain language and get answers with their sources.
- Answers based only on your own documents, using RAG
- Local AI models on your own hardware when data must stay in-house
- A clear "I don't know" when the documents hold no answer
Workflow automation and AI agents
Your team stays in charge of decisions, while AI workflows handle repeatable steps such as checking, sorting or preparing information. Tools such as n8n can connect these steps to your existing systems.
- Clear rules and human approval at every important step
- Connections to your existing tools, documents and inboxes
- Visible usage, so costs and results stay easy to follow
Prompt engineering and AI skills for teams
Employees get better results from AI when they know how to ask. I help teams write clear requests, so everyday work with AI stays professional, consistent and efficient.
- Prompt guidelines written for your company's tasks and tone
- A shared prompt library for recurring work
- Custom AI skills for specific roles and tasks
Document and data processing
Emails, invoices and contracts hold facts your systems need. AI reads them, extracts the key details and passes them on, while a person reviews unclear cases.
- Key facts passed into your existing systems
- Human review for unclear or unusual documents
- Clear rules for what AI may decide and what it may not
AI advisory and cost control
Not every project needs a build. I give an honest second opinion on where AI helps, where it does not and which approach fits your budget.
- Local or cloud AI models: which fits your data
- Where token costs come from, and how to keep them low
- Data risks reviewed before any build starts
How we work together
From the first conversation to handover, the process stays simple and transparent. We can work online, in person, or a mix of both.
- First conversation. We talk about your goals, your tools and your data.
- Proposal. You receive a clear approach with scope, timeline and costs.
- Build in short steps. Regular updates keep progress visible.
- Handover and support. The solution goes live, with support if you need it.
Talk it through
Tell me what you are trying to solve. We will look at it together, with no pressure.