Skip to content
@oughtinc

Ought

Ought is a product-driven research lab, developing mechanisms for delegating open-ended thinking to advanced machine learning systems.

Our projects include:

  • 🔬 Elicit, the AI Research Assistant
  • 🧊 ICE, a trace visualizer for language model programs
  • 📓 Factored Cognition Primer, a tutorial on how to write compositional language model programs

We are always looking for talented and motivated people to join us or collaborate with us. If you are interested, please check out our careers page.

You can also follow us on Twitter to stay updated on our progress.

Pinned Loading

  1. ice ice Public

    Interactive Composition Explorer: a debugger for compositional language model programs

    Python 535 66

  2. primer primer Public

    Factored Cognition Primer: How to write compositional language model programs

    48 7

  3. ergo ergo Public

    A Python library for integrating model-based and judgmental forecasting

    Jupyter Notebook 108 23

  4. mosaic mosaic Public

    A web app for recursive question-answering with pointers

    TypeScript 37 3

  5. patchwork patchwork Public

    Command-line recursive question-answering with immutable contexts and explicit data store

    Python 24 1

  6. raft-baselines raft-baselines Public

    Python 13 10

Repositories

Showing 10 of 35 repositories