Edwards-Alexander built a 150,000-line application, roughly 120 kLoC of it Rust, without reading it: "I didn't read or review any of the code, except occasionally, out of interest." Its data access layer reached 17,155 lines in one file. He…
In the News: August 1, 2026, Morning
A Thoughtworks CTO measured what refactoring saves an agent, and got 83% fewer input tokens for the same change.
Morning edition
Machine-readable
Download Markdown
Story
Refactoring an agent-written codebase cut the token cost of the same change by 83%
Read story →
Story
A factory that survived its own failures, and the instrument it used to find the next one
A queue bug double-counted PR comments, so two sandboxes would race to conflicting commits; it was fixed and regressed across roughly sixty PRs in two or three days. What held was not another patch.
Story
Two cross-harness cost benchmarks, and they disagree
Same model (GPT 5.6 Sol at medium), same environment, same agentic tasks with known-good outcomes, harness versions pinned. API token totals ran from 172,807 for smol to 5,073,137 for Claude Code, a 29.4x spread with the model held…
Story
Y Combinator open-sourced the agent harness it runs its own operations on
The core accepts interchangeable harnesses, "pick your own harness and model and switch between them," with Pi, OpenCode, Codex and Claude Code all driving it through one wiring file. Its predeclared command policy, hard denials for…
Also this cycle
Permalink
[Hacker News](https://news.ycombinator.com/item?id=49135257) at 207 points and 93 comments at 3.8 hours, read 15:12 EDT 1 August
Cursor removed cost information from the usage page and CSV export
A Cursor staffer called the CSV break accidental and the graph removal deliberate, because included-plan usage in dollars "is not what you're billed." On the vendor forum a staffer defended it as "deliberate design," then reversed the position roughly a day later according to a contemporaneous capture. The forum thread could not be reopened independently, so that reversal remains unverified here.
Everyone is building LLM routers, we deprecated ours
A first-party negative result on a shipped harness component, bounded in time and population: launched March, deprecated June, off for good on 1 September, "after four months of usage across 7000 cloud users." A pre-loop router cannot see what it is routing on, because "the prompt alone does not contain the whole task; it is just the trigger." Manifest's own team, arguing against a feature they shipped.
Thread watch
Permalink
Hacker News, 113 points and 94 comments at 6.3 hours, read 15:12 EDT 1 August
On the non-use of AI in my writing process
Charles Stross has written about refusing AI in his own work. The post was not read, so the figures describe the discussion only. Whether it belongs in this feed is open: we have covered a software project's formal AI contribution policy before, and a novelist's personal practice may not be the same object.