I find where AI and automation pay off in your business, build it, and maintain it.
Worked with
Pick a part of the business. Adjust the steps to match how you work, set your team’s hours, and bring the number to our first call.
≈ 710 ha year back. About 18 full-time weeks.
Send me this estimateYour hours × 46 working weeks × each step’s share of the time. An agent step counts 80 %, an assisted step 40 %. It’s a starting point, and I check it against real data before anything is built.
Find.
I sit with the people who do the work and map where the hours go. You get a ranked list of where AI pays, what it costs to build, and what should stay human. You keep the map whether or not we build it together.About the AI assessment
Build.
I build it myself, in your stack: agents, integrations and the interfaces your team will use, with tests, monitoring and the guardrails regulated businesses need.About AI developmentAI agentsAI automation
Run.
I host and run it for you: monitoring, security, updates and steady improvements. If you would rather run it yourselves, your team gets the code, the docs and the training. Your choice, and no lock-in.About hosting and maintenance
One scored audit of everything you ship: code, sites, apps, APIs, containers, infrastructure, cloud and mobile builds. It runs from the terminal, in CI or from an AI agent over MCP.
Buyers describe what they want once, draw their areas on a map and get matched every day when the right home is listed. Sellers list for free, with or without an agent.
My own agent harness, where a crew of AI agents works in my projects. Every request, tool call, file change, cost and approval is visible while it happens. It is how I build, and you get the same visibility in what I build for you.
Read the casePrivate project, in daily use
Also: Lingo, an AI-native localization platform, and contributions to T4 Stack, an open-source starter for typed web and native apps.
I have been programming for more than ten years and building digital products for web, iOS and Android professionally since 2019. I studied computer engineering at Chalmers, led cross-platform development as tech lead at weapp and worked as a developer at Telia.
Today I build AI solutions for companies and run my own platforms, SökHem and OpenAudit, from architecture to operations. In the world’s largest React chat community I am recognised as a star helper.
| 2024– | AI consultant and developer | Anton Wester AB |
| 2022–2024 | Tech lead, cross-platform | weapp |
| 2021–2022 | Software engineer | weapp |
| 2019–2021 | Software developer | Telia |
| 2016–2019 | B.Sc. computer engineering | Chalmers |
I build in TypeScript, React, React Native and Node/Bun, run Postgres, Redis and Docker in production, and integrate language models from Anthropic, OpenAI and others with agents, tools and MCP.
It depends on how much of the business it covers. After a first call you get a fixed price for the assessment, so you know the cost before we start.
A contained first step, such as an agent that sorts and answers incoming email, can often be running within a few weeks. Larger integrations take longer, and the assessment shows how long.
I build with data minimisation and clear logs, choose providers and regions by where your data may be processed, and document which decisions AI makes and which a person approves. That makes GDPR and the AI Act easier to meet.
Whatever fits the task. I work with models from Anthropic, OpenAI and others, and with open models that can run on your own server, and I build agents that use your existing systems through APIs and MCP.
Either works. By default I host and run the solution for you, with monitoring, security updates and steady improvements. If you would rather take over, your team gets the code, the docs and the training.
I am based in Gothenburg and happy to work on site in western Sweden, for example during the assessment. The rest is usually remote, and I take on work across Sweden.