Posts
Newest first · Stack map →
Aug 18, 2026
The agent platform, described in one place for once
I have been building an agent platform on my own hardware one post at a time. This is the whole thing in one place: how a job gets dispatched, which model runs it, and what it has to prove before I believe it.
Aug 1, 2026
My agents' instructions were a dozen copies with no owner
The instruction text my AI tools read lived in a dozen places across the fleet: files on each machine, symlinks, importers, and text boxes in someone else's web UI, with no record of which was current. Now there is one registry that owns every surface by path and hash, a checker that verifies 103 of them, and a ruling about what an embedded version token actually means.
Jul 28, 2026
Letting whichever machine is free do the work
Jobs were queuing behind a busy machine for no better reason than habit about which machine owned that kind of work, so I separated who runs a job from where the job lands.
Jul 27, 2026
Two background workers, one status column, and a system that never converged
Prospect's two AI background workers were quietly overwriting each other's status, so every service restart redid work that was already done. The handoff called it a one-row cosmetic cleanup. Reading the two workers against each other showed something worse, and the fix cost no GPU time.
Jul 27, 2026
Letting the fleet choose which AI model runs each job
I stopped picking which AI model runs each job by hand and built a router that scores the options on what each one has left, reserves the winner, and writes down why it chose.
Jul 20, 2026
The file I was re-reading on every single turn, and a handoff I treated as a claim
Two things about working with AI agents that only show up after you have run them for a while: the cost of what you read at the start of a conversation is paid again on every later turn, and a detailed handoff from another agent is a claim until something the author could not have shaped confirms it.
Jul 15, 2026
One conversation, one section: how I keep months of AI-assisted work from losing the thread
Working with AI assistants for months across several projects, the failure I kept hitting was not bad answers, it was lost context: every conversation restarted from zero or dragged a bloated history. The fix was a method: each conversation does exactly one bounded section, and continuity lives in three small documents, not in the chat.
Jul 5, 2026
Handing a job to whichever machine should run it, then walking away
I stopped doing my machines' work at the keyboard and built a way to hand a task to whichever machine should run it, let it run on its own, and pick up the result later.