AI for businesses
I add AI to existing workflows, such as internal search and handling enquiries or documents. First we scope the job and test it against real examples. The solution logs what it does, so you can see when a person needs to take over.
- I usually reply the same day
- Written estimate before work starts

star reviews
AI
The work ranges from customer service chatbots that answer from your own documentation, through agents that handle recurring casework, to product copy that is generated and maintained automatically. What all of it has in common is that the output is measured, logged and can be switched off.
The jobs in demand right now are integrating language models into internal tools, setting up RAG on your own data, automating email and document handling, and reviewing pilot projects that worked in the demo and fail in production. If you are starting from scratch, we begin by finding the process where the payoff is biggest.
The audience is companies that want AI to solve a concrete, well-defined piece of work. You own everything I build: prompts, code, test data and documentation live in your own accounts, so you can change supplier without starting over.
The work comes from all over Denmark plus a few clients abroad. I am based in Glostrup outside Copenhagen and work remotely with companies in Aarhus, Odense, Aalborg and the rest of the country, with workshops and kick-off meetings on site when it makes sense.
How businesses can use artificial intelligence
In practice, AI for businesses is about fairly ordinary work: the tasks where someone reads, sorts, writes or looks for something many times a week. Artificial intelligence in the form of language models is good at exactly that. It can understand text that never looks the same twice and turn it into something a system or a colleague can use.
In customer service, AI can sort incoming email, find the answer in your own documentation and earlier cases, and prepare a draft for someone on the team. In administration and finance it is usually about invoices, order confirmations and contracts arriving as PDFs. AI extracts the details and enters them into the accounting system, so nobody has to key them in by hand.
In sales and marketing, AI can pull together a customer's history from the CRM and email before a meeting, write the first draft of a proposal, or keep product copy and translations up to date across many SKUs. Internally, staff can look things up in handbooks, procedures and product data and get an answer with a source, instead of asking the colleague who always knows.
What the good examples have in common is that the task happens often and follows a pattern, and that a person approves where it matters. A reply to a customer or a payment should be seen by someone before it moves on. The rest can often run on its own, as long as everything is logged and the cases the system is unsure about land with a person.
How to get started with AI
AI for businesses does not have to begin with a big project. What works is starting small and concrete.
- Start with what you already have. If you have Microsoft 365 or Google Workspace, AI features often come with the subscription, and some of your staff may already be using ChatGPT or similar on their own.
- Ask the people doing the work where the time goes. They know which tasks repeat every week, and which ones just feel tedious.
- Pick one task rather than an overall strategy. One well-defined task quickly shows whether AI helps. A plan for the whole company easily ends up as a document nobody uses.
- Test on your own examples, not on a demo. At the same time, write down how long the task takes today, so you can see afterwards whether it helped.
- Decide up front what a person always has to see. Typically that is anything going out to a customer or moving money.
- Agree the ground rules before usage spreads: which tools may be used, and which data must never go in. More on that below.
AI policy and the rules
If your staff use AI, you should have an AI policy, even if you have not built anything yourselves. Without one, people use the tools anyway, just without anyone having agreed how. It does not need to be long, but it should answer three things: which tools may be used, which data must never go in, and who approves new tools. Personal data, confidential customer information and trade secrets, for example, do not belong in a free tool you have no agreement with.
The EU AI Act entered into force in August 2024 and is being phased in over several years. The requirements depend on what AI is used for and how high the risk is. Most everyday uses, such as assistants, drafts and search in your own documents, sit at the low end, where the requirements are few. One requirement does apply to every company using AI at work: since February 2025 you have had to take steps to build AI understanding among the staff who use the tools, including what the tools can do and where they can go wrong. This is known as AI literacy.
If you use AI for decisions about people, for example in recruitment, employee assessment or credit scoring, it may fall under what the regulation calls high-risk, where the requirements are far greater. Get legal advice before you start there. And regardless of the AI Act, GDPR applies as always when personal data is involved. You need to know where the data ends up and have a data processing agreement with the vendor. I am not a lawyer, but I build solutions so personal data is kept out wherever possible.
Specific services
AI development that reaches production. I have built software since 2010 and worked on 400+ projects. Tell me about your idea and get a reply the same day.
Read moreAI implementation that gets used day to day. One area at a time, measured results, built by a developer with 16 years of experience. Get a same-day reply.
Read moreAI consultant who builds the solutions himself. Honest advice on where AI pays off in your business, based on 16 years as a developer. Get a same-day reply.
Read moreAI automation of routine work with documents, email and data. Built by a developer with 400+ projects behind him. Describe the task and get a same-day reply.
Read more
How it works
Four steps. You know what happens when, and you can stop after any of them.
- 01 / 04
Pick the process
Together we find the piece of work that repeats often enough for automation to pay off. The more contained the task, the faster the first version reaches production.
- 02 / 04
Set a baseline
We record how long the task takes today, how often it occurs, and how many errors slip through. Without those numbers, nobody can tell afterwards whether the solution was worth the effort.
- 03 / 04
Build against a test set
I collect real examples from your day-to-day and test every prompt and model change against them. That keeps quality stable when the solution is tuned later on.
- 04 / 04
Production with logs and a kill switch
The solution is wired into your existing systems with full logging and a switch that can turn it off. I document the setup and train someone on your side to maintain it.
Client reviews
Brands I've worked with
Frequently asked questions
What is the difference between AI and generative AI?
AI, artificial intelligence, is the umbrella term and also covers technology that has been in use for many years, such as spam filters, webshop recommendations and forecasts. Generative AI is the branch that creates new content, such as text, images or code, based on patterns in huge amounts of data. ChatGPT, Copilot and Claude are examples. When companies talk about AI today, they usually mean generative AI. Although the name is about creating, in a business it is at least as useful for reading, sorting and extracting information from text.
Do we need an AI policy?
Yes, if your staff use AI tools, and many already do on their own. An AI policy sets out your internal ground rules: which tools are approved, which data must never go in, what has to be checked before an answer is used, and who to ask when in doubt. Keep it short enough that people read it, and update it when you adopt new tools. I help write it as part of AI implementation.
What does the EU AI Act mean for an ordinary business?
For most, it comes down to a few concrete things. First and foremost, you have to take steps so that staff understand the AI tools they work with. That requirement has applied since February 2025, and it is easier to meet when you have an overview of which tools are used and for what. If your customers talk to a chatbot, it should be clear that it is AI. If you use AI for decisions about people, for example in hiring or credit scoring, the requirements may be far greater, and you should get legal advice. The Danish Agency for Digital Government (Digitaliseringsstyrelsen), which coordinates supervision in Denmark, has published guidance for businesses.
Which AI service should we start with?
It depends on where you stand. If you do not know where AI can pay off, or you want a vendor proposal assessed, AI consultant is the right place to start. If you have licences for ChatGPT or Copilot but they are used at random, AI implementation helps you bring AI into one area at a time. Routine work with emails, PDFs and data moved by hand belongs under AI automation. If a case has to be investigated across several systems, that is a job for AI agents, and if you have a concrete idea or a prototype that needs to hold up in production, it is AI development. If your staff need to get better at the tools themselves, the AI workshop under Courses & workshops is the obvious place. If in doubt, write and tell me where you stand, and you will get a reply the same day.
Your next step
Get a quote for AI
Tell me about your project. I’ll get back to you with questions, options and a concrete next step.