Dogfood Release
Pre-release
Pre-release
Features:
- Chat interface for the user to enter the problem statement. The user can interact with Sirji by providing change requests and enhancements.
- Message protocol implemented for various agents - Planning Agent, Coding Agent, Research Agent, and Executor Agent.
- The Planning Agent: Takes in a problem statement and breaks it down into steps.
- The Coding Agent: Sequentially generates code to solve the problem programmatically.
- The Research Agent: Utilizes RAG and gets trained on URLs and search terms. Later, it can use the training knowledge to answer the questions asked by the Coding Agent.
- The Executor Agent: Responsible for creating the code files, running commands, and installing dependencies.
The Planning Agent and Coding Agent utilize OpenAI Chat Completions API, but they can be easily extended to infer from other LLMs.
Similarly, the Research Agent currently is built using the OpenAI Assistants API and can be easily extended to use vector databases to store the embeddings and implement the RAG approach.
You can see what’s possible with the dogfood release in the demo videos: https://github.com/sirji-ai/sirji/blob/v0.0.1/demos/README.md