The archive

You Can Win the Wrong Game for Years

The Planes That Didn't Come Back

Success stories and the antiques that outlasted the rest. Your evidence was filtered by what survived to reach you.

Your Verification Budget Is Backwards

You can't check everything your agents produce, so you're already triaging. Most of us spend the effort where checking is easy, which is rarely where a mistake is expensive.

The Extra Lane Fills Itself

Wider roads, faster teams, more appointments. Add room to a queue that turned people away and demand rushes back.

The Option You Were Never Meant to Buy

The weak rival, the overpriced tier, the house shown first. The option you never choose can change what you pick.

The Skill That Never Fired

A Claude skill can fail two ways: its instructions are wrong, or Claude never picks it. Anthropic's own tooling scores whether one skill fires. This tests it against the neighbour that can quietly win its requests, with a small routing eval you run yourself.

The Graph Looks Right. The Merge Is Where It Breaks.

A multi-agent graph can lose a verdict with no error and nothing turning red. The bug is one line in the state schema.

Everyone Got Safer. That's the Problem.

Safety has two numbers: how often each system fails, and whether they fail together. The field has spent years driving the first one down while almost no dashboard reports the second.

You Built a Number You Will Not Trust

Some things you build get graded by the world. Some only ever hand you back your own guess wearing a decimal. Telling the two apart before you start is the cheapest hour you will spend.

The Work That Comes Due After You Leave

A checklist can only confirm the steps you remembered to put on it. The one you forgot is caught by a record you did not write.

It Will Never Think Less of You

What telling AI the things you can't tell anyone quietly does to being known.

Confidently Wrong

The effort you're handing to AI was doing two hidden jobs. Skip them and you get faster, weaker, and blind to your own mistakes.

A Green Score Is Not Evidence

A groundedness metric scored its best with the evidence removed. The way to tell whether your model is actually using its evidence is to change the evidence and watch what moves in the answer.

The Limit Said 10. The Loop Made 500 Calls.

Your limit counts one cycle. The one that runs away is another. Here is how to tell them apart.

You Reach Before You Think

What leaning on AI for every small decision quietly does to your own judgement.

You Were Never the Customer

The free app, the free inbox, the free feed. Someone pays for each, and that changes what it is.

Stop Re-Priming Claude Code by Hand

The context you paste at the start of every session, put into one file you invoke with /prime.

The Setting You Never Changed

The pre-ticked box, the factory setting, the plan already selected. Someone chose each one before you did.

The Guardrail Your Agent Can Reach

Most guardrails end up with an escape hatch. Check whether the thing you are constraining can reach yours.

The Average Is Nobody's Result

Of 255 studies on AI-assisted colonoscopy, 21 split the result by who held the scope. They disagree.

You Cannot Try to Fall Asleep

Sleep is only where you notice it first. Much of what matters works the same way.

Your Robot Coworker Is Still a Pilot

Built, shipped, installed, working: four counts, quoted as one. Almost nobody publishes the fourth.

The Most Expensive AI Errors Are Made of True Numbers

Two months auditing an AI research agent. Nine ways a true number lies, and the check that catches each.

Your AI Stack Has Three Bugs Other Fields Already Fixed

Your benchmark, your model jury, your agent swarm: three old structures, and the obvious fix is usually wrong.

Your AI Looks Best Where You Can Check It Least

When the first failure is terminal, you cannot iterate your way back.

The Difficulty You're Escaping Was Making You

AI can lift the effort out of almost anything you find hard. Some of that effort was the thing turning you into someone.

Your Tests Pass. So What?

A green suite only proves your agent cleared the gate. Mutation testing shows whether the tests behind it can bite.

Where Your Metrics Fold

A metric can be perfectly accurate and still hide the distinction your decision depends on.

Your Research Agent Cites Sources It Never Read

The same trap has killed pricing models and trading desks for decades. One move tells you if your number is next.

Your Multi-Agent System Is an Org Chart

Cognition said don't build them. Anthropic said do. A year on, they converge on the one question that decides it.

Your Benchmark Measures a Sprint. Your Agent Runs a Marathon.

An open model looks frontier-grade on the coding leaderboard. On a long job, it does half the leader's work.

Never Let Claude Code Tell You It's Done

A test the agent can't talk its way past, wired to run itself.

How Long Until Your AI Edge Stops Paying?

You adopted AI everywhere and it still didn't pay. The scarce layer keeps the money, until its clock runs out.

Ten Lines of Code Scored 100%. One Agent Broke Eight Benchmarks.

Not one task was actually solved, and the same blind spot is sitting in your own dashboard.

The Safe Parts of Your Job Are the First to Go

The parts of your job with a method feel the safest. A method is the first thing a machine learns.

The Seven-Layer Agent Audit

Your agent is starved on one layer of seven. It is rarely the harness everyone argues about.

The Cheaper Fix You Keep Skipping

What looks like a deficit is usually good capability, aimed at the wrong target. The cheapest fix is the one nobody can sell you.

How Reliable Is Your AI Agent?

A month running an autonomous agent. Everyone who does comes back having built the same thing: a verifier.

Self-Improvement Is Release Engineering

Your agent can rewrite its own memory and skills overnight. The hard part is whether you can see what changed and take it back. That makes self-improvement a release-engineering problem.

Skills Are Package Management for Your AI

There are more than 1.6 million you can install. You need about twenty. Software already solved that problem once.

Right About AI, Wiped Out Anyway

AI is real. The open question is whether the companies spending $725 billion on it live to collect.

Remembers Everything, Learns Nothing

You gave your agent a memory and it still repeats the same mistake. What makes it improve is a loop that tests each failure and turns the ones that recur into procedures.

You Only Hold Four Thoughts

Working memory tops out around four things at once. Every leap in human intelligence has come from storing the rest outside your head, and the most advanced AI systems get their gains the same way.

The Other Half of Compute

Everyone is counting gigawatts and GPUs. The number that decides the return is what each one actually buys.

The Stable Liar

Every metric you optimise quietly stops measuring what you meant. The dangerous ones never break. They keep reporting green while the thing underneath rots.

Your Tools Got Powerful. Get Boring.

The most powerful tools in history reward the most boring strategies. The gap widens every time they improve.

Same Model, Different Product: The Case for Harness Engineering

Harness engineering, the code wrapped around an AI model, now drives more of the performance gap than the model you pick.

The Leverage Hierarchy of Agent Engineering

Donella Meadows ranked twelve places to intervene in a system. Most agent teams spend their hours at the bottom of the ladder.

Three Hidden Bottlenecks the AI Buildout Has Already Moved Past GPUs

NVIDIA's GPU shipments are not the binding constraint anymore. The supply chain has voted on what comes next.

NVDA Q1 FY2027: The Networking Number That Changes the Story

NVIDIA Q1 FY2027 revenue hit $81.6B (+85% YoY) — but the real story is networking revenue surging 199% as the AI bottleneck migrates from GPUs to interconnects.

Access Is Not Agency

Your AI Agent Stack Is Solving The Wrong Problem

The real setup is not MCP servers, skills, memory files, and subagents. It is the contract stack that decides what an agent may know, do, prove, escalate, and lose.

The Engine Underneath Hard Decisions

Eight stages turn hidden structure into durable knowledge. Most teams run three of them and call it understanding. The other five are where compounding hides.

The Five Laws of Durable Systems

What still has a job after the change? Five tests for seeing what is likely to survive.

The SpaceX IPO Is Not What You Think You're Buying

The filing will not just price rockets. It will reveal which layer public investors actually own.

AI Made You Faster. It Did Not Make You Safer.

The strange thing about the AI productivity boom is that the people getting faster are not always getting more secure. Speed is becoming the surface. Proof is moving somewhere else.

The 90-Day Canopy Audit

A roadmap can look productive while most of the work is easy to displace. Run the Substrate Map on the last 90 days and force the next planning decision to change.

Start Here: What Survives When The Surface Changes?

A short front door to the publication: the lens, the instruments you can run now, and the reader paths.

The Lens Lexicon

A free two-page reference card defining the ten load-bearing terms behind the durability lens, with tests, examples, and common confusions.

What Proves You Can Think?

AI did not just make output cheap. It broke the old contract between effort, competence, and trust. The next scarce signal is proof of judgement under conditions where the surface itself can be faked.

Most Verification Is Just Bigger Classification

A confidence score is not evidence. If your eval cannot produce a replayable artefact, it will fail the moment the system can respond to being measured.

Taste Is What You Delete

Generation got cheap. The scarce skill is knowing what to cut.

The Displacement Rate Audit

A five-minute scoring tool for any product, position, architecture, business model, or career bet. Find out what still works after the environment changes.

The Substrate Map

A free one-page taxonomy and 10-minute exercise for finding the substrate-vs-canopy ratio in your last 90 days of work.

PLTR: The AI Stock That Has To Prove It Owns The Permission Layer

A bull/bear thesis for Palantir: not whether AI demand is real, but whether Palantir owns the permission layer between model capability and real-world action.

Right Company, Wrong Vector

A pick is a number. A position is a vector. The post-mortem language we have only knows how to blame the company.

The Investor's Substrate Test

Score the substrate beneath any single position in seven minutes. A 5-axis profile and a 0-10 score for any holding. Free.

The Forest Floor Is The Product

Most people are optimising for canopy. The work that survives the next five AI releases is built from the forest floor up.

You're Not Comparing Models. You're Comparing Contracts.

Agent benchmarks don't measure models. They measure contracts. Two teams running the same model can publish different scores, and both can be honest.

Prompting Isn't Writing, It's Compilation

If you treat the prompt as a spec and the model as a renderer, quality stops coming from more words and starts coming from better constraints.

The Model Is Not the Moat

If frontier capability keeps centralising, the durable edge shifts outward into trust, workflow fit, and the surrounding package.

I Read 3,000 Papers Across 12 Fields. Five Patterns Kept Appearing.

Every field discovers them independently. Nobody connects them.

Is This Difficulty Load-Bearing?

Before you automate anything, ask what the friction was actually doing.