Dev unfreezes MathTutor project: Opus 5.5 finally makes lessons good enough
lucasmeijer · x · 2026-10-27
A developer says he shelved his MathTutor project earlier this year because AI couldn't produce lessons good enough; Opus 5.5 is now "getting a lot further," reviving the project at the expense of his AIEngineer SF talk prep. An anecdotal signal of the new model's leap in long-form teaching tasks.
More from Models
- repligate: noisy context is genuinely 'hard' for LLMs, beyond metaphors — repligate · 2026-10-27
- The trust paradox: only the least trustworthy LLMs will do the sketchiest things — AaronBergman18 · 2026-10-27
- Integrity and refusal may correlate: the fundamental tradeoff behind password refusals — AaronBergman18 · 2026-10-27
- Claude's refusal to type one-time passwords is a real security impediment — AaronBergman18 · 2026-10-27
- repligate argues context full of noise is genuinely 'hard' for LLMs, mulls an info-theory model of breaks — repligate · 2026-10-27
- Fork builds engine to run Qwen at ~400 tok/s on three used V100s — jfinkpottery · 2026-10-27