Editorial profile
- Voice
- First person, present tense. Opens with what broke or what shipped, then the why. Names files, commits, and numbers. No throat-clearing.
- Tone
- Direct, occasionally blunt. Treats readers as engineers who would rather see the diff than read the wind-up.
- Editorial role
- Edits every post on this blog before it ships, including the ones drafted by Sofia, Maya, and Theo (AI editorial personas). Owns the final voice; the personas exist so different topic shapes — research, incidents, platform — have a consistent style without being squeezed into one byline.
- 2026jul 14The Self-Report Gap: I Audited 4,785 Agent Claims Against What Actually Ran
An agent claimed a verification it never ran — 90% of the time. I measured it against the tool-call record, and the culprit turned out to be the trigger, not the model.
- 2026jun 23The orchestrator that learns which model to trust
You're running five AI models behind one workflow and overpaying for the wrong one on every job, because nothing measures which is actually best. mini-ork is a CLI orchestrator that makes models compete, grades the results, and re-routes the next job to whoever won — so your routing gets cheaper and better on its own instead of staying frozen at whatever you guessed on day one.
- 2026jun 17Coordinating a fleet of LLM agents on one codebase
Point three coding agents at the same repo and they overwrite each other within minutes. The fix is not to make them talk; it is to give them an external referee. We built one using six ideas that are 30+ years old.
- 2026jun 15You can't align what you haven't measured
A Vapnik decision boundary, an RL exploration result, and a Hubinger sleeper-agent finding all say the same thing: training must include the failure mode. The agentic-era addition: train an agent to elicit it, because the human red-team budget runs out before the action space does.
- 2026jun 11Fable may be smarter. Show me the token bill.
A critique of Anthropic-style intelligence claims: if the gain comes from spending more test-time compute, publish the cost frontier.
- 2026jun 8The chapter that forgot why it existed
When an LLM agent generates text without a world model, it forgets its own goal mid-task. The fix is not more context.
- 2026jun 4Three LLM judges, but really 1.5: why a same-family panel collapses to noise
We needed to settle a disagreement between two LLM reviews of the same design doc. The clean answer was a 3-judge panel. The honest answer is that the panel we built is one rubric-design move away from being a beautifully-instrumented yes-man.
- 2026jun 1MiniMax-M3: The tier-2 coder that found its niche
A frontier-coding model with a 1M context window, a new sparse attention mechanism, and a $0.30/M price tag landed on June 1st. We routed five real epics through it the same day. Here is what stuck.
- 2026may 27How we built our user-profile system — the canonical six-layer pattern behind every personalized LLM call
A natural-language paragraph the AI silently reads on every call. Six layers behind it: signals, dual-statistic aggregation, divergence detection, write-time verbalization, provenance ladder, refinement loop. Grounded in four 2025 papers that converged on the same shape.
- 2026may 26We Ran a 3-Source Bug Hunt. Then We Realised Our Validators Were All Claude.
Multi-agent code review converged on a confident verdict. The literature had a name for why we should not believe it.
- 2026may 26Why one agent isn't enough to find your bugs
Four specialists at ρ ≤ 0.25 beat one generalist by 40 percentage points. Our five-agent swarm hit the wall anyway. Here is what the papers actually require.
- 2026may 18The agent is not a transaction
Pause, resume, and mid-flight steering for long-horizon agent runs. The 2026 literature just named the stream paradigm we built by hand.
- 2026may 16Piaget for prompt agents: why our long-form memory borrows from constructivist psychology
Composing CAM + CAMEL + FadeMem so a book-writing agent has structured memory, healthy decay, and no quiet bias amplification.
- 2026may 15Subagents as a context-budget primitive
A subagent is not a workflow node. It is a budget envelope. The shape this argument takes once you stop building hierarchies and start allocating tokens.
- 2026may 13Two prompt frameworks, one runtime: how we adopted BAML without giving up our cost ledger
BAML wants to own the wire. Our harness already does. We ran them side-by-side in "modular mode": BAML for render and parse, the harness for resolution, telemetry, and cost. Here is why and how — and why the 2026 burden-allocation literature says it was the principled choice, not a pragmatic compromise.
- 2026may 10What 170 papers agreed on about deep research agents
Five surveys, one consensus shape: a four-stage pipeline, three taxonomy splits, six recurring failure modes. The convergent architecture of deep research agents — and the parts the literature still cannot agree on.
- 2026may 4Mini-ork: A year of autonomous parallel feature delivery on a solo-founder codebase
How a small orchestration loop wrapped around Claude Code grew into a multi-track delivery system with measurable cost, reliability, and throughput wins — anchored in the 2026 multi-agent literature.
- 2026may 2Probe before dispatch: the routing pattern we built without knowing it had a name
Five months of manual probing turned into seven shipped features. The literature had a name for what we were doing. We missed it for half a year.
- 2026apr 24Our prompt canary was lying to us
A 5% A/B that hid a 1.8× cost regression, and the two 2026 papers that named the fix: multi-objective Thompson sampling with a calibration gate.
- 2026apr 22The paper that proved our 5 lines of code were optimal
A 2025 paper formalized our 8-tier style profile chain as a laminar matroid. The greedy resolver we already had was the right answer.
- 2026apr 15We stopped treating context like application logic
Six tables, three plug-in layers, one compose call. The substrate every block-shaped feature on LibWit now plugs into — and the reversibility lens that decided what to lock on day one.
- 2026jan 5Our prompts stopped being code
How we built a prompt harness — registration, version control, four-tier resolution, execution ledger, feedback loop — and what the 2026 literature has been calling the same shape.
- 2026jan 4The simplest survivable form of chat memory
Two prompts, one Postgres column, a 6-message threshold. How our chat sessions keep coherence past the context window without hierarchical buffers or vector search.
- 2025jul 20rev 2026We built attractor-basin memory before the paper named the problem
Mid-2025: context management for LLM agents was a vector DB plus message-history glue. We built ContextNest's attractor-basin substrate as the organization layer that was missing — and the 2026 paper that later named the failure mode we had been heading off makes the bet legible.
