Sweet PalNotes from the people building an app for type 1 diabetes: artificial intelligence, vibe coding and everything in between.
An AI assistant for type 1 diabetes can observe, show and explain, but must never decide a dose. The safety rules we built into Sweet Pal, and why.
Posted by Simone Tolomelli at · Labels: T1D · 6 min read
AI chat cost lives in the tokens a model reads, not the ones it writes. Multiple rounds, prompt caching, one model per role and the real cost per subscriber.
Posted by Simone Tolomelli at · Labels: AI · 6 min read
A coding agent starts from zero every session. How a handoff file, read first and rewritten last, keeps it on track on the same product for weeks.
Posted by Simone Tolomelli at · Labels: Vibe coding · 5 min read
How we do LLM evaluation before release: reference sentences against the real model, versions switched off but never deleted, and time to first token.
Posted by Simone Tolomelli at · Labels: AI · 7 min read
Vibe coding on a real product only holds up when the rules live in files the agent rereads every session. Ours, one by one, and the reason behind each.
Posted by Simone Tolomelli at · Labels: Vibe coding · 5 min read
We took the deciding away from the LLM: JOLs classify in parallel, a script writes the brief, the model only speaks. Sweet Pal now replies in about 1.5 s.
Posted by Simone Tolomelli at · Labels: AI · 7 min read
