Tutorio

Slop is cheap, understanding is everything.

Tutorio is a tool designed to help developers onboard themselves to unfamiliar repositories. Unlike traditional documentation, it educates users through interactive learning, guiding them through the process of creating a specific feature they would like to build within the codebase.

Project Structure

The project is divided into two main components:

• backend: A Python-based FastAPI server that orchestrates AI agents to generate personalized lesson plans and grade interactive exercises.
• nexhacks: A Next.js frontend application providing the user interface for project selection, context gathering, and interactive lesson consumption.

Core Philosophy

Tutorio focuses on building deep understanding rather than generating superficial code. It leverages AI to analyze an existing repository and construct a structured series of educational stages, including conceptual explanations, comprehension questions, and practical coding tasks.

External Tools & Technologies

This project utilizes the following tools, libraries, and technologies:

AI & Agent Orchestration

• Google ADK: Agent development kit for building internal agent logic.
• LiteLLM: For cross-model and cross-provider LLM support.
• Arize Phoenix: LLM orchestration observability and tracing.
• Seda AI
• Gemini CLI
• Antigravity
• Swarm.dev

Backend Stack

• FastAPI: Core Python web framework.
• Loguru: Structured logging for backend operations.
• Uvicorn: ASGI server for running the FastAPI application.

Frontend Stack

• Next.js: React framework for the user interface.
• React Markdown: Rendering tutorial content with full markdown support.
• Simple Git: Client-side and server-side git automation.
• Tailwind CSS: Utility-first styling.
• Lucide React: Icon system for the user interface.