An agent built a Tauri desktop application end to end. The pure logic was correct on the first or second attempt. Every defect that mattered sat at a boundary no headless test can reach.
7 min read
A scanned NASA specification, a novel, the memory of a product, and a sealed algorithm. What four controlled rebuilds establish about where AI-assisted reconstruction of legacy systems actually fails.
8 min read
A compressor reconstructed from behaviour alone, graded against an answer key sealed until the work was frozen. The decoder scored 100 percent and the encoder 8.3 percent on the same format.
7 min read
HAL/S runs again: twelve preserved Shuttle-era programs, verified value-for-value against an independent interpreter. The load-bearing human work was refusing a majority vote between OCR engines.
6 min read
Capitalisation rewards crossing from exploration into development, so the boundary quietly rewards early commitment to a design. An incentive on technical decisions nobody describes as technical.
5 min read
Familiar means familiar to the receiving team, not old. And the ledger records what the code cannot do, including the debt nobody chose.
6 min read
The system can be rebuilt; the encoded definition of working cannot. Ship the harness first, and test it by making a deliberately wrong answer fail.
4 min read
Whether prototype code survives is a decision per component, made in advance, and separate from whether the learning survives. Neither always-rewrite nor demo-implies-retain holds.
5 min read
Three patterns that build a thin end-to-end path before breadth. Two of them overlap and the third is a different kind of thing, because it has to keep an incumbent running alongside.
5 min read
Tell us what you're working on and where you need help.