About
I am a backend engineer who moved into AI. Based in Oslo, Norway.
What I do
Currently at Izy I work on both halves of the stack. The production backend serving Norwegian commercial buildings, and the AI running on it. Computer vision self-checkout for workplace canteens, and demand forecasting so kitchens buy the right amount of food. Both in production, both in daily use.
Keeping a heavy system running is a different discipline from building a model. Doing both is most of what I have to offer.
Eight years across backend architecture, machine learning and technical product, from startups to enterprise. Distributed services, data pipelines, ETL, and systems that had to stay up as they grew. Most of it before the current AI cycle began.
Why ethics is not a side interest
The gap between a model that works and a system you can defend is where most AI projects quietly fail. I spent 160 hours studying AI ethics and fairness formally through Turing College. Fairness metrics, causal interventions, governance, production deployment.
One idea from it changed how I think. The impossibility theorem says you cannot satisfy every reasonable definition of fairness at once. So “is this system fair?” has no purely technical answer. Somebody has to choose, and they should choose deliberately. I wrote about that in Coding Is Solved. The Hard Part Starts Now.
Research
- Anomaly-based Intrusion Detection with Machine Learning
- Blockchain-powered Supply Chain (IJAE, Vol. 2 No. 2, 2019)
Also a Registered Scrum Master.
Technical background
Python, AWS, microservice architectures, ETL pipelines, machine learning, and a range of backend frameworks. Enough breadth to have opinions about which parts of a stack deserve the complexity they carry.
Outside the work
Global geopolitics, the long arc of human civilisation, and psychology. Sailing when the weather in Norway allows it. All of it is about how systems behave once people are inside them.