Behind the curtain
ShuttleAI is comprised of multiple LLM agents, working together to handle your request. Starting from the beginning itself – they expand upon the user’s prompt, reason about the project and define a plan of action. While giving a detailed prompt works the best, even simpler prompts like “Build me a todo app” will be enough for the agents to make assumptions and define the basic shape of the app.
Once we have our plan decided, we pass it on to the code-generation agents that focus on generating you a working codebase. Meanwhile, our Shuttlify agents take care of your infrastructure and provisioning, while our compilation agents sit in the background and fix any potential errors or mistakes that might appear. What you get in the end, is a real, live Rust-based backend deployed on Shuttle’s cloud.