Articles
Longer arguments from the research programme, with their sources.
Entries marked ↗ live on another site, named under each; the others are pages of this site.
- Chaos engineering for behaviour
Red-teaming a model's behaviour is chaos engineering, and operations already wrote the rules for it.
- Knowledge Infrastructure for LLMs
A language model can only work from what someone wrote down and put in front of it. Most companies that bought AI maintain their code and little else.
- The Stance Layer Is Still Toil
A hook can block a word on every reply. Nothing I run can yet catch a model before it folds, only after.
- The cheating moved
A reading of the Goodhart Labs chess honeypot through the stance frame, three additions to the design (an escalate grade, observer-invariance, cold and warm arms), and a dated, falsifiable prediction.
- An RCA on Claudish
Where Claude's writing style came from: text written to be heard, read in silence.
- Which LLM User Are We Talking About?
Advice about language models depends on which model, on whose hardware, inside how much harness.
- Compaction Is the New OOM
Most agent harnesses ship without swap.
- What Operations Already Knows About Running Agents
Error budgets, reconciliation loops, separation of duties and recovery over prevention: four operations practices that fit agent work almost line for line.
- Write for the Codec
Nobody reads your documentation. Both ends are a model now, and the file between them is a wire format.
- The Golem Made of English and the Horizon of Consequences
A language model is a golem made of English. It behaves well only where it can see what its act will cost, and infrastructure is the craft of bringing that cost into view.
- The Trap File Is Longer Than the Instruction File
Seven months of running an agent-built pipeline unattended, 308 lines of recorded failures next to 208 lines of instructions.
- Sycophancy Is Layered
A longitudinal single-user case study, with operationalized artifacts.