Context Clues
How to keep AI agents building what matters, not just what compiles.
- No. 71IKEA EffectBuilding something yourself and that thing actually being good are two different facts.7 min
- No. 70Interrater ReliabilityA shared label in a shared field doesn’t mean two people are making the same judgment when they choose it.7 min
- No. 69Revenue RecognitionCash arriving and a service actually being delivered are two different facts about the same contract.6 min
- No. 68Near-DecomposabilityThe same structure that makes a specialist good at their specialty is the structure that makes certain real connections invisible from inside it.6 min
- No. 67Situation AwarenessKnowing every fact that happened this week and knowing where things stand are two different mental operations.7 min
- No. 66Costly SignalA claim any competitor could make for free carries no information about which one is true.7 min
- No. 65Ecological FallacyA property true of a group doesn’t transfer to a member of it just because the group is real and the math behind it is correct.7 min
- No. 64SoundnessA plan can read as a single coherent story from goal to task and still not be true of the company running it.9 min
- No. 63State EstimateA backlog is a measurement.9 min
- No. 62Revealed PreferenceWhat a user says and what a user does aren’t two versions of the same fact.7 min
- No. 61ThroughputStory points measure a guess made at the start of the sprint.7 min
- No. 60DriftA crawler can tell you a page is different from last time it looked.6 min
- No. 57Entity ResolutionThe same decision shows up under three different names in three different tools.6 min
- No. 55Proactive BriefEvery tool that holds what your team knows waits to be asked.7 min
- No. 54Two AnswersGive a coding agent the same task twice and it hands you two different, plausible answers.6 min
- No. 53Greek FireThe Byzantines held a weapon their enemies could barely counter, and then lost it entirely, because the one thing that mattered was never written down.7 min
- No. 52Sixteen AgentsBrief looks like a single assistant.6 min
- No. 51StigmergyTermites build coherent nests with no blueprint and no foreman, coordinating through the traces they leave in a shared environment.8 min
- No. 50Working MemoryAn agent runs almost entirely on working memory.7 min
- No. 49Overruled PrecedentMatching the code around a change is right almost every time.9 min
- No. 48Seven SecondsEvery plank of the ship gets replaced, so what makes it the same company is memory.12 min
- No. 47Why coding agents ignore your rulesYou wrote the rule down and the agent broke it anyway. It is not defiance, and a longer file makes it worse.10 min