Rewriting campfire in Jai: developer finally feels "models are amazing when tasks have perfect verification"
lucasmeijer · x · 2026-10-17
lucasmeijer shares hands-on experience: using models to rewrite a campfire effect in Jai was the first time he truly felt in his bones that "if your task has perfect verification, models are amazing." Previously he found models useful but not great at new "I'll-know-it-when-I-see-it" work — echoing his earlier point about the ocean of difference between large work with a reference implementation and designing something new.
More from AGI Musings
- Having your agent read 8,000-word articles may be the future of serious online writing — danshipper · 2026-10-17
- AI productivity hype meets reality: the last 20% of a chart took 80% more time — vboykis · 2026-10-17
- French math community clash: stop mourning AI, Chinese mathematicians use it fully — IgorCarron · 2026-10-17
- Millions of cloud-resident AI agents could trigger a compute demand surge — Thionne_WTZ · 2026-10-17
- AI-written complaint emails make ESOL clients unintentionally rude, trainer warns of backlash — Sense_Difficult · 2026-10-17
- Dev's take: every new AI benchmark — StarCraft, GUIs, Minecraft, CAD — is secretly a coding benchmark — wateriscoding · 2026-10-17