I’m an independent IT consultant with nearly 17 years of experience. I’ve spent most of that time building systems that make other engineers’ lives easier.
I started as a web developer in 2009 — ASP.NET, then Ruby on Rails, then the full frontend parade: jQuery, Backbone.js, Ember.js, Angular. After a few years of that, I made the leap to infrastructure and never looked back.
The evolution
Once I discovered automation, there was no going back. Ansible for configuration management. Terraform for cloud infrastructure. Docker when it was still a curiosity. Kubernetes before it was boring. I’ve worked with EKS, AKS, and GKE, and across all three major cloud providers — Azure, AWS, and GCP — as well as smaller ones like DigitalOcean, Scaleway, Linode, and Hetzner.
The way I work
I’m not afraid to get my hands dirty. I like figuring things out. I treat my spare time like a personal research lab — playing with new technologies for the fun of it. That knowledge has a habit of paying off in unexpected places.
Take Docker: I started tinkering in my spare time, gave a talk at a meetup, and ended up at Container Solutions embedding with Google’s PSO team in Amsterdam and Zurich. Or networking: I set up a FreeBSD router out of curiosity, moved to OpenWRT, learned about VLANs, switched to OPNsense — and later that knowledge turned up in a client project at exactly the right moment.
That’s the pattern. Learn something because it’s interesting. Ship it when it matters.
The IoT thread
I’ve been running Home Assistant for years, experimenting with ESP8266 and ESP32 boards. At a hackathon, a friend and I built a snow-depth detector using an ultrasonic sensor and an ESP32, shipping metrics over MQTT. The organisers were impressed enough to hire me to turn it into a real product.
More recently, I built a hardware demo kit for 84codes — connecting DHT11 sensors to LavinMQ message queues, with a Ruby backend forwarding data to Grafana dashboards. It made their conference workshops tangible, and that was the point.
Right now
I’ve been running large language models locally on a server I built specifically for that purpose. Not because anyone asked me to — because I’m curious about what happens when you take that capability offline.
That’s what I do. I learn things for the fun of it. Sometimes the knowledge turns into something useful. Sometimes it doesn’t, and that’s fine too.