IdeaWell

Project ideas from Hacker News discussions.

Postgres rewritten in Rust, now passing 100% of the Postgres regression tests

Original Article

Hacker News Discussion

📝 Discussion Summary (Click to expand)

1. AI‑generated code quality is dubious and needs verification

"Most of the code on the internet is already a black box to you." — colordrops
"If you have a black box that spits out code, and you are unable to distinguish the quality between a top tier dev and an AI inside the black box, then the distinction is unnecessary." — dwedge

2. Rigorous regression testing is essential to trust a rewrite

"Regression tests are exactly production scars: every time you fix a bug, you add a non‑regression test to ensure you don’t break it again." — hvb2

3. Maintenance and expertise are major hurdles for large rewrites

"The maintainers that wrote those tests will have experience you won’t get out of a rewrite." — rustyhancock

4. Performance gains drive architectural changes in AI rewrites

"The new version passes 100% of the Postgres regression suite, uses a thread‑per‑connection model, and is 50% faster on transaction workloads and ~300× faster on analytical workloads." — malisper

5. Licensing and broader implications are part of the debate

"If you don’t like the license just let an LLM spend a few days “porting” it and give that port any license you like because that is apparently what we do now." — _tom____

🚀 Project Ideas

[LLM Code Quality Scoreboard]

Summary

  • Provide an automated, reproducible quality score for AI‑generated code to help teams distinguish reliable rewrites from “slop”.
  • Quantify risk via test pass rate, static‑analysis metrics, and code complexity, enabling data‑driven adoption decisions.

Details

Key Value
Target Audience Engineering teams using AI coding assistants
Core Feature Auto‑generated quality score and risk assessment for AI‑generated code
Tech Stack Python backend, Rust static analyzers, GraphQL API, Docker containers
Difficulty Medium
Monetization Hobby

Notes

  • HN commenters repeatedly lament “slop” and would welcome a quick, objective metric.
  • Could be integrated into CI pipelines, sparking discussion on acceptable quality thresholds.

[Regression Test Auto‑Synthesizer for Rewrites]

Summary

  • Teams struggle to expand test coverage for AI rewrites; manual test creation is costly and error‑prone.
  • Generate additional regression tests by fuzzing inputs and comparing original vs. AI‑generated outputs, surfacing edge‑case gaps.

Details

Key Value
Target Audience Database maintainers, language‑porting projects
Core Feature Comparative testing and gap detection between original and AI‑rewritten code
Tech Stack Go fuzzing engine, PostgreSQL test suite, Web UI for result visualization
Difficulty High
Monetization Hobby

Notes

  • Directly addresses HN concerns about “unnecessary friction” and “slop”.
  • Serves as a Building block for safer AI rewrites and invites community contributions.

[License Compliance Analyzer for AI‑Transpiled Code]

Summary

  • AI transpilation often blurs license boundaries, risking inadvertent violations.
  • A CLI tool that scans codebases, identifies source licenses, and automatically generates SPDX metadata and attribution files for AI‑derived portions.

Details

Key Value
Target Audience Open‑source maintainers, legal teams
Core Feature License detection, provenance tracking, auto‑generated SPDX metadata
Tech Stack Node.js, Licensee library, SQLite database for provenance
Difficulty Low
Monetization Hobby

Notes

  • HN users frequently raise licensing worries; an automated compliance check would be highly valued.
  • Can be packaged as a GitHub Action, encouraging widespread adoption and discussion.

[Unsafe Block Linter for Rust Ports of Legacy Systems]

Summary

  • Rewrites preserve hidden unsafe patterns from original C code, leading to subtle bugs.
  • VS Code extension that highlights unsafe blocks, suggests safer Rust idioms, and visualizes provenance of unsafe sections.

Details

Key Value
Target Audience Rust developers porting C/C++ legacy code with AI assistance
Core Feature Linting of unsafe usage, refactor suggestions, lineage graph of unsafe blocks
Tech Stack Rust, Tree‑sitter parser, React UI, Language Server Protocol
Difficulty Medium
Monetization Hobby

Notes

  • Addresses HN’s focus on memory safety and the “unsafe” concerns around AI rewrites.
  • Could become a mainstream linter plugin, attracting safety‑focused contributors.

[Distributed Database Benchmark Suite for AI Rewrites]

Summary

  • Performance claims of AI rewrites are often vague; the community needs reproducible benchmarks.
  • A benchmark framework that runs standardized TPC‑C and analytical queries across multiple DB ports, automatically generating comparison reports and visualizations.

Details

Key Value
Target Audience DB engineers, performance analysts, startup founders
Core Feature Multi‑engine benchmark runner, result dashboards, regression tracking
Tech Stack Python, ClickHouse, Grafana, Docker‑Compose for isolated services
Difficulty Medium
Monetization Revenue-ready: Subscription tier for enterprise analytics

Notes

  • Directly tackles HN debates about performance vs. memory safety, providing empirical data.
  • Has clear commercial potential, encouraging paid enterprise usage and sustained development.

Read Later