Opinions — stdout

01 ~/posts/underspecified.md
The Specification Spectrum: Narrowing the Guesswork

// AI code generation is really a specification problem — and a spectrum of tests, types, and formal methods that narrow the guess.
#languages #correctness #verification Jun 18, 2026 7 min read →
02 ~/posts/fuzzing-llms.md
The Jagged Neurosymbolic Frontier

// Invariant inference, execution oracles, and the limits of pure AI audits: how a "poor man's neurosymbolic" loop forces LLMs to write down their assumptions, then fuzzes the assumptions themselves.
#fuzzing #llms #security Jun 19, 2026 13 min read draft
03 ~/drafts/vuln-window.md
Implications of the Collapsing Vulnerability Discovery Window

// As AI compresses the time from disclosure to working exploit, the patch-window math every security team relies on stops holding.
#security #disclosure not yet published draft