Project Hearthmind
Hearthmind Software
The inference and orchestration stack that actually runs on the Spark day to day.
Progress
0% complete
Objective
What this quest sets out to do
Select, configure, and document the software stack — model runtime, orchestration, and household-facing interface — that turns the Spark from idle hardware into a system the household uses daily.
Why it matters
Why Solystopia cares
Hardware ownership without a usable software layer is a $4,699 space heater. This quest is what makes the machine legible and useful to a non-specialist household member, not just to the person who built it.
Components
What this is actually made of
Milestones
Where this quest stands
2025 Q4
Ollama runtime installed and benchmarked
Done2026 Q1
Household-facing chat interface selected
In_progress2026 Q1
Model library finalised for 128GB ceiling
Planned2026 Q2
Vector database and observability layer integrated
Completes the minimal rebuildable stack described in the evidence base's section 9.2 research priorities.
PlannedOpen design questions
What is not yet decided
- Should the household interface be a simple local web UI, or does it need to work from any device on Hearthnet?
- Which model sizes actually deliver a good day-to-day experience inside the 128GB memory ceiling, once tested rather than assumed?
- How much of the stack's ongoing maintenance burden can realistically be automated away versus requiring continued manual attention?