This year I retired from a lifelong career in software development. It was a dichotomy: I was tired of the task lists and meetings and schedules and difficult people — but at the same time I was having more fun making software than I ever had in my life, because of AI. I'm one of those developers who doesn't fear being replaced by it (it can't, I'm retired; if I were a recent grad I might feel differently) — to the contrary, I'm grateful that code generators want to do the boring part of my job: writing the actual code. I couldn't just walk away during this moment, so I quit my job but didn't quit making systems.

Growing cannabis is a hobby of mine, and in retirement I want to take the time to refine the craft. By now the same realization I had is probably hitting you: I could use AI to grow weed. There are sites with AI baked in that help this way — but they're in the cloud, and you have to upload your data before they can process it.

So I decided to learn Edge AI and build a device that analyzes my grow in real time without ever loading images or anything else to the internet. (I live in a state where growing is legal, though still not at the federal level.) It's a small box that works only on the local LAN, reads from cameras and sensors inside the tent, and runs an ongoing analysis — deficiencies, pests, disease, stress — and reacts to events like seedling→veg and the flip to 12/12, dispensing advice I put together from the published literature — a controlled-environment cannabis textbook and the peer-reviewed papers around it, each claim kept next to the source it came from.

Running the first grow through it now. It didn't take over the world, but I did have lots of corrections for it. It did hold the air right on target while the plant got sick anyway — which is the part I'd want to talk about. Happy to get into the on-device stack.