Thistlebridge
A house in Utah with a greenhouse, some servers, and questions about what comes next.
A marginal property — awkward parcel, freeway noise, not picturesque. The greenhouse produces plant starts for the neighborhood — tomatoes, peppers, herbs, flowers through the growing season. In a back room, servers run AI models locally, without cloud dependencies. The two are connected: we're testing whether local AI infrastructure can support practical skill development without replacing human capability.
This is a home where ordinary domestic life happens. The experimental work fits around that, not the other way around.
What We're Testing
Voice documentation while working in the greenhouse. You're learning through doing, hands occupied. Instead of fumbling with voice memos or losing thoughts, you talk to the system like you'd talk to a colleague. It transcribes, organizes, connects new observations to previous work.
A narrow test — but if natural language can mediate between human intention and technological capability during physical work, it suggests something larger might be possible.
What's Here
At the north end of the dome, a pond holds a blue lotus (Nymphaea caerulea). The wicking system on the shelves can start 1,080 plants at once. The servers in the back room run on repurposed enterprise hardware — about $2,000–3,000 total for capable local AI.
We're documenting what we learn: what works, what doesn't, what questions emerge.
If you're interested in local AI infrastructure, practical skill development, or the intersection of the two, you might find something here. If not, that's fine too. We're not trying to build an audience. We're trying to figure something out and sharing the process for whoever finds it useful.
Questions? Try asking a Thistlebridge-specific Claude instance here.
This site was developed with LLM assistance. We're testing practices, not just describing them.