We should be more ambitious with the software we build
- ID
- 39238
- Status
- summarized
- Published
- 25 Oct 2026, 5:47 AM
- Fetched
- 25 Oct 2026, 11:48 AM
- Provider
- Hacker News
- Category
- dev-community
- Original URL
- https://danluu.com/ambitious/
- Source URL
- https://hnrss.org/best
Summary
- Score
- 7.5
- Created
- 25 Oct 2026, 11:48 AM
- Tags
- Audience
- developersvibe_codersdatabase_learnerssaas_startup_founders
What happened
The essay argues builders should be more ambitious with software, using Michael Malis's pgrust project as an example of something that can look like AI-generated slop but has surfaced real Postgres bugs through thorough testing. It cites bugs in SP-GiST operations, JSONPath results, pgcrypto, the dictionary loader, and money overflow, while acknowledging most rewrite projects are low-quality and often tout slop benchmarks. The Hacker News thread drew 242 points and 204 comments.
Why it matters
If you are evaluating an ambitious rewrite or AI-assisted project, the concrete signal here is upstream bug discovery and testing depth, not the performance claim; pgrust's testing reportedly found real Postgres bugs in SP-GiST, JSONPath, pgcrypto, the dictionary loader, and money overflow. If you are building one, budget for rigorous testing and benchmarking before making performance claims, because the essay's example shows that thorough testing can produce durable value even if the project itself fails.
Discussion angle
What evidence would make you take an ambitious or AI-assisted rewrite seriously: upstream bug reports, reproducible benchmarks, or code review? And what minimum testing and benchmarking bar should a team set before shipping or praising one?