Rewrites.bio: 60x speedup in Genomics QC and AI rewrite principles for Science
emiller88
15 points
3 comments
April 02, 2026
Related Discussions
Found 5 related stories in 87.6ms across 8,303 title embeddings via pgvector HNSW
- We rewrote JSONata with AI in a day, saved $500k/year cjlm · 94 pts · March 26, 2026 · 51% similar
- TurboQuant: Redefining AI efficiency with extreme compression ray__ · 509 pts · March 25, 2026 · 50% similar
- GPT‑Rosalind for life sciences research babelfish · 83 pts · April 16, 2026 · 50% similar
- AlphaEvolve: Gemini-powered coding agent scaling impact across fields berlianta · 274 pts · May 07, 2026 · 49% similar
- Be sinceerly human – AI to undo your AI writing arnon · 11 pts · April 23, 2026 · 48% similar
Discussion Highlights (3 comments)
emiller88
AI coding assistants have made it possible for domain experts to rewrite established scientific software in days. We believe that a wave of AI-driven tool rewrites is coming to bioinformatics. We've published a set of best-practices principles to help people to approach rewrites in the right way. Along the way we fully rewrote and open-sourced the genomics QC tools for RNAseq, the most widely used genomics pipeline, yielding a >60x performance improvement.
tenzin12
Really interesting direction. The validation and drop-in compatibility part is what caught my attention most. If this holds up across more datasets, this could remove a huge amount of RNA-seq QC runtime pain.
teekert
So needed. I just threw genebody_coverage and TIN out of RSeQC ;) Maybe they can come back now.