Developer finds LLMs excel when tasks have reference implementations or perfect verification

Developer lucasmeijer observes a sharp capability divide: LLMs excel at large tasks with reference implementations or perfect verification criteria, as shown by his Jai rewrite of a campfire effect, but struggle with open-ended novel design.

2026-10-17 ~ 2026-10-17 · 2 related posts

• Dev observation: models excel with a reference implementation, but struggle at designing something truly new — lucasmeijer · 2026-10-17
• Rewriting campfire in Jai: developer finally feels "models are amazing when tasks have perfect verification" — lucasmeijer · 2026-10-17