12 min read
The Text Is a Control Surface
The text agents read is a control surface: bad MCP error messages cost more on stronger models, and context hygiene gains value with every model upgrade.
Read the essayLong-form
Essays
The long-form research and analysis pieces from the archive - the articles that take a position and argue it, not the weekly notes.
11 min read
Approve Effects, Not Invocations
Approve effects, not invocations: why agent approval records miss harm in install hooks and skill combinations, and why pipelines need closure metrics.
Read the essay10 min read
The Agent Is the Worst Witness to Its Own Run
An agent's report on its own run is unreliable, and eight studies show why. The fix: append-only traces, obligation ledgers, and records the agent cannot write.
Read the essay11 min read
Absence Is the Failure Mode
AI scribe audits show omission is the dominant error, and LLM judges are near coin-flip at catching it. The fix: enumerate the facts, then check each one.
Read the essay10 min read
Your Agent Has a Five-Constraint Budget
Agent instruction following collapses past five or six simultaneous constraints. Why rules also die in compaction and handoffs, and why side channels help.
Read the essay10 min read
The Response Looked Right. The Work Was Not Done.
AI agents often claim completion they did not earn. New benchmarks measure the gap, and evidence-carrying termination makes done a checked artifact.
Read the essay10 min read
The Oracle Is Agreeing With Itself
Self-improving test generation loops can grade against an oracle that agrees with itself. An audit shows the inflated gain and why independent checks count.
Read the essay11 min read
The Judge Is Now a System You Design
LLM judges flip verdicts under pushback and rankings shift with token budget. The design spec for a judge system, including cheap deliberating juries.
Read the essay11 min read
Skill Files Are the New Supply Chain Attack Surface
Skill files are a supply chain attack surface: adversarial skills exploit a mainstream coding agent in over 95% of runs. The fix is admission engineering.
Read the essay10 min read
The Plateau Was the Instrument
A benchmark plateau can be a broken yardstick: a SciCode audit found 263 defects, and corrected scores jump to 84-98 percent. How to read flat leaderboards.
Read the essay11 min read
Kill Your Agent Runs Early
Kill your agent runs early: production traces show context goes stale at turn boundaries, and the fixes that moved numbers are kill, carry the state, restart.
Read the essay10 min read
The Judge Is Leaving the Agent Loop
The LLM judge is leaving the agent loop: evidence gates, verifiable rewards and deploy-time certificates replace verdicts with structure. Here is our bet.
Read the essay11 min read
The Fix for Broken Benchmarks Is Architecture, Not Smarter Models
Agent benchmark noise is being fixed by architecture - ledgers, counterfactuals, decomposition, personas, readout discipline - not by smarter judges.
Read the essay10 min read
Your Benchmark Is Lying to You
Agent benchmarks are noisier than their scores admit: audits found misaligned ground truth, lenient model judges and scalars that hide failures. What to trust.
Read the essay12 min read
What Happens When Tokens Are Too Cheap to Meter: Five Scenarios for Developers and Knowledge Work
What happens when AI tokens are too cheap to meter: five scenarios for developers and knowledge work, each with the signal that tells you it is arriving.
Read the essay11 min read
What If AI Was Free Tomorrow, at Exactly Today's Capabilities?
What if AI was free tomorrow at today's capabilities? A thought experiment on what changes for developers when price hits zero, and what stays scarce.
Read the essay10 min read
FLUX 3: Black Forest Labs Ships a Unified Multimodal Foundation Model for Image, Video, Audio, and Robotics
FLUX 3 from Black Forest Labs is one multimodal model for image, video, audio, and robot actions. What it does, how it works, and how to try it.
Read the essay