Dev complaint: models are still horrible at code reuse
snowclipsed · x · 2026-10-09
Developer snowclipsed notes that AI models remain terrible at code reuse — instead of identifying and reusing existing functions and abstractions in a codebase, they tend to rewrite implementations from scratch. The short take resonated as a real, under-discussed weakness of current coding models.
More from Models
- Anthropic impresses on efficiency but users feel "babysat" by its guardrails — kimmonismus · 2026-10-09
- COLM paper: "forks in the road" in post-training data shrink reasoning model coverage — ParshinShojaee · 2026-10-09
- Google's open medical VLM MedGemma publishes in Nature Medicine — ymatias · 2026-10-09
- Researcher asks: do model providers' ToS allow training on business/API customer content via synthetic data? — RylanSchaeffer · 2026-10-09
- LFM 2.5 5.4B seen as better laptop pick; 8B A1B lags 2.6B dense — Aggravating-Push-207 · 2026-10-09
- Users Report Claude's $800 API Credit Refund Expires Just 10 Days After Claiming — Daniel_Farinax · 2026-10-09