Loading the edition…
· Past edition · 9 stories
Retrospective edition reconstructed from later captures. Paper versions may include later updates; coverage labels describe the evidence used.
Applications & products
The HN capture carries a secondhand benchmark report: performance reportedly rose from 0.31% to 88.94% of theoretical ceiling in roughly 40 hours; the linked article was not fetched.
Why it matters: If Ho’s quoted repeatability claim holds, improving benchmark software after first silicon could shorten the interval before production ramp-up and shift schedule assumptions. This capture cannot verify the outcome.
Applications & products
Open each story in the reader. Summaries are shown when available; some sources provide an abstract only. With an account you can keep your own notes.
The story title reports that Claude Code reads AGENTS.md when Claude.md is absent. The captured thread treats this as a compatibility step, while disputing what it fixes and why it arrived.
Why it matters: For people sharing project guidance across coding tools, the reported fallback could reduce the need for a Claude-specific file. But comments distinguish finding AGENTS.md from reliably following its instructions, so teams may still need to check both behaviors.