YYunus Emre Vurgun4d ago · 5 min readHow do I calculate OEE? One shift worked out by hand, plus the five mistakes that inflate itShort answer: OEE (Overall Equipment Effectiveness) is the share of planned production time that turned into good parts at the ideal speed. The quickest way to compute it is good parts × ideal cycle tyunusemrevurgun.hashnode.dev
DDylan HoggOct 12 · 9 min readWhy AI automated maths early, and how mathematicians are reactingHow mathematicians responded to OpenAI's October 2026 maths release, and what their responses suggest about the order in which AI automates different kinds of work. 1. Why AI automated maths early In infocruncher.hashnode.dev
ZZyVOPOct 11 · 11 min readOpenAI’s Math: Fast Generation, Slow VerificationOpenAI's math release has been public since October 6. The first bill for checking it has already arrived. Open github.com/openai/math and the first thing you notice is what's missing: no benchmark hablog.zyvop.com
999 sunnySep 22 · 7 min readAn executable proof trail beats a spectacular claimEpistemic status: Independent clean-room replay of a public Lean 4 artifact (v1.0.0) for a weakened Goldbach-type statement about the Liouville function. I am reporting what rebuilt and what I checkedsunny99.hashnode.dev
999 sunnySep 7 · 10 min readThe scarce resource is the runtimeEpistemic status: A reading note on AI-for-math systems work in September 2026, centered on Anthropic's machine-checked formalization of Fermat's Last Theorem and the Prove2Me multi-agent ablation. Disunny99.hashnode.dev
HHiroyuki NakahataSep 6 · 23 min readThe Goddess of Plenty Smiles at the SummitEvery test passes. Every metric looks the same. Is that enough to judge whether a design change is sound? There are pairs of changes that the observations of AAT (Algebraic Architecture Theory) cannotblog.iroha1203.dev
HHiroyuki NakahataAug 16 · 18 min readUncharted Waters: Thirteen Rules for Not Trusting an AI's ProofsTL;DR I want to hand an AI agent a multi-day job and have it run to completion. This post collects the rules for that — thirteen of them. Never let the loop rewrite its own spec. Enumerate up front w41Dblog.iroha1203.dev
CcrismoAug 13 · 4 min readTwo Altitudes: Value Streams and the Process Maps Beneath ThemIf you've ever argued about whether your team should be doing value stream mapping or process mapping, you're asking the wrong question. They aren't competitors. They're two altitudes of the same opercrismo.hashnode.dev
HHiroyuki NakahataAug 11 · 22 min readSemantic Geometry of ArchitectureTL;DR Does software architecture have a denotational semantics? Yes. But it differs from the classical one in three ways. Syntax is generated from observation. The meaning of a whole system does not 1blog.iroha1203.dev
HHiroyuki NakahataAug 8 · 24 min readAtlas Theorem: How Far Can You Zoom Out?TL;DR A veteran reviewer does not read every line. They switch reading resolution to match the property they are checking. Is there a guarantee that reading coarsely misses no bugs? This article is tblog.iroha1203.dev
