I build AI systems that solve business problems.
Three years building with AI: first inside a large consultancy, now for my own clients. Before that, years on the business side, which is why I start from the problem and not from the model.
MSc Civil Engineering, eMBA, three years deep in AI. I design the solution and build it end to end, from the business need to the thing running in production.
From business need to full production. Listen, understand, build.
At NTT DATA I ran enterprise AI delivery: an agentic system that summarises clinical files for second medical opinions, and an AI automation solution for customer service support at a logistics operator. Since April 2026 I have been building solutions for my own clients from Zürich: knowledge graphs over contract bases, regulatory radars, SAP-integrated automations, among other things.
What I like is owning a project end to end: working side by side with the business people who will actually use it, agreeing on what good looks like, and delivering something that gets adopted and moves a number. A system nobody uses is not a solution, however good the model behind it is.
What I have done.
Two in production: SAP-integrated bill of materials, generated straight from the client order, for a custom joinery manufacturer. And a contract management platform, a Karpathy-style wiki with AI at its core, for a large Colombian logistics group.
One large build in progress: Automated test reporting for one of the top national engineering firms in the United States.
Smaller work: Regulatory radars and other legal-tech AI solutions.
Biggest result: +40% productivity across a team of 175 customer service agents at a logistics operator.
Also delivered: A €150k agentic-AI project leading a team of four, and a €170k technology transformation shipped in six months.
Ownership: Full commercial, risk and contractual responsibility on every engagement.
I speak at industry conferences and universities about applying AI to real business processes.
Real systems, not demos.
(Happy to walk through any of these in detail.)
Wiki, knowledge graph, draft generator and alerting over the contract base of a Colombian logistics group.
Better retrieval, so the system can answer more ambitious questions against the whole contract corpus.
AI automation for customer service support at a logistics operator. +40% productivity across a team of 175 agents.
Builds the test report from the initial test design and the machines' raw data: charts and written conclusions, delivered as finished artifacts. For one of the top national engineering firms in the United States.
Ingests and summarises large volumes of clinical files so physicians can issue a second opinion. Team of four, seven months.
Generates the bill of materials for custom joinery orders, integrated into SAP and into the machine production orders.
Parameterising the rest of the company's product range so more order types are covered.
A personal project: we build and train our own assistants to run our digital life, from a second brain to expenses and finances, calendar, tasks and notes. Everything runs on our own hardware and stays private.
Tracks executive appointments and M&A across Europe's largest companies, and raises an alert where a legal practice may have a commercial opportunity.
Filters 15 Latin American authorities and scores the regulatory impact on the sector.
Looking for a team to build with.
Hiring someone to build the AI solutions your business actually needs? Write to me.
I read everything myself.