Typed results for raw SQL in TypeScript, with no codegen or live database
Emerged finds new products worth a look, daily — what’s new today
If you're tired of hand-writing JSON schemas and tool wrappers, Nola derives them from your TypeScript types at compile time.
Nola extends TypeScript with LLM inference as syntax, modelled on async and await. An infer function defines an inference context, an Intent<T> holds the data and instructions for the model, and ask resolves it into a typed result. JSON schemas come straight from your types, dot-prefixed parameters mark what the model can see, and native functions can be called as tools. It compiles to plain TypeScript. There is a VS Code extension, a local trace console, and an offline playground. Apache-2.0, at v0.1.7.
Typed results for raw SQL in TypeScript, with no codegen or live database
A small local model that picks from your options and says how sure it is
Open-source DSL that freezes repeated agent work into typed, inspectable steps
A coding harness that pairs with you instead of working on its own
Hardens your shipped builds so AI reverse-engineering agents waste effort
On-device AI autocomplete for almost every text field on a Mac