Jonathan Nye
I'm an engineer who builds software for factories. Originally from South Africa, I now live in Berlin.
Work
I work at Juna AI, where we build AI agents for factories. Before joining Juna AI, I worked at Tesla, developing software for manufacturing during the Giga Berlin ramp. I have also worked in real estate, healthcare, and energy.
My background is electrical engineering, but most of my career has involved data, machine learning, and software. I like problems where the software has to deal with the real world. Factories provide plenty of those.
Why “Uncommon sense”
In my first year at university, one lecturer often answered a question with: “But that's just common sense.” It usually wasn't.
Somehow that stayed with me. It only becomes common sense once you've done the work of understanding something.
This blog is where I (sometimes attempt to) write down some of that work. It is mostly about engineering, manufacturing, AI, energy, and whatever else I happen to be working on.
Research
My earlier research covered renewable generation on electricity networks and using machine learning to understand energy demand.
- Customer Segmentation Using Unsupervised Learning on Daily Energy Load Profiles — Journal of Advances in Information Technology, 2016.
- Generation increase on distribution feeders using electronic voltage regulators — IEEE PES T&D, 2014.
- Local reactive power droop control modification for distributed generators — IEEE PES T&D, 2014.
- Reactive power control of embedded generation on radial distribution networks — PowerGen, 2014. Publication listing on LinkedIn.
- Design and implementation of a variable speed wind turbine emulator — International Conference on Electrical Machines, 2012.
Bushwak
I also built Bushwak with my wife. It's a social safari app for African parks, available on iPhone and Android. It helps you identify wildlife, keep a record of your sightings and photos, and learn more about the animals you encounter.
Offline use is a central part of the app. You can record sightings without a connection, then share them when you're back online. Species guides and a map of sightings bring together the field guide and the record of your trip in one place.
Some posts
- Production vibe coding — my view on AI coding and the price of good work.
- Plotly Dash is all you need — why I think Dash is enough for most analytics projects.
- Introduction to OEE — an explanation of the manufacturing metric I use most.