Inspiration
I wanted to make codebase analysis more understandable than a simple list of warnings
What it does
Code Archaeologist identifies useful patterns, architecture layers, outdated code, and technical debt.
How we built it
We used Python, FastAPI, Tree-sitter, NVIDIA Nemotron 3 Ultra, Nebius Token Factory, and an interactive 3D frontend.
Challenges we ran into
We faced deployment difficulties because integrating and deploying Nemotron-generated code required careful dependency, API, and hosting configuration.
Accomplishments that we're proud of
We built a working analysis tool with CLI, API, reports, model routing, and a 3D interactive demo.
What we learned
AI works better when it receives structured information about a codebase instead of raw source code alone.
What's next for Code-archaeologist
We plan to support more languages, improve live repository analysis, and add deeper team collaboration features.
Built With
- fastapi
- html/css/javascript
- nebius-token-factory
- nvidia-nemotron-3-ultra
- python
- tree-sitter
Log in or sign up for Devpost to join the conversation.