DotPrompt
RESEARCH LOG

Paper history

LLMs Are the Runtime

The original paper treated the model as a runtime and the prompt as source code. It proposed compiling a .prompt file into an intermediate representation and an execution plan for a particular model.

Creating Reliable Systems with Imperfect Models

The current thought experiment follows an account-isolation requirement through discovery, repair, and release. Its examples show how a review agent can connect a change to the evidence a release decision requires.

Download current PDF ↓