This project is an attempt to fully integrate LLMs and other forms of "Artificial Intelligence" into the creative process of building software. The goal is to aid developers in implementing features, fixing bugs, testing, and documenting their code in a fast and reliable manner.
By leveraging existing state-of-the-art models, distributed computing, and the ability to read your local codebase, we aim to generate code that is both syntactically and semantically correct, fully tested, and well documented.
An iterative process of a feedback loop between the developer and several AI actors should allow for a fine-tuned experience that is tailored to fit whatever the project needs to reach its goals.
This is day 1 of the project. We are actively planning it's architecture, deciding the scope of the initial MVP, and setting a roadmap. We are also looking for contributors to help us build this project.
We are looking for contributors with a wide range of skills. If you are interested in contributing, just create an issue or a pull request and we will get back to you as soon as possible.
Any and all ideas are welcome, so don't be shy! We are especially interested on developers with experience in Elixir, Phoenix, and Bumblebee.