Inspiration
What it does
How we built it
Challenges we ran into
Accomplishments that we're proud of
What we learned
What's next for IssuePilot
Inspiration
We wanted to create something that could help developers spend less time finding the cause of errors and more time fixing them. This led us to build IssueAgent using NVIDIA Nemotron through Nebius Token Factory.
What it does
IssueAgent can analyse individual pieces of code or public GitHub repositories. It identifies possible errors, explains the problem, shows affected lines and suggests fixes. It can also monitor registered repositories using GitHub Webhooks.
How we built it
We built IssueAgent in Python using Flask for the web interface, gitingest for reading GitHub repositories and NVIDIA Nemotron 3 Super 120B through Nebius Token Factory for the AI analysis.
Challenges we ran into
Our main challenges were moving the project from a terminal program to a web application, processing full GitHub repositories, configuring the Nebius API and connecting GitHub Webhooks to the analysis system.
Accomplishments that we're proud of
We are proud that we managed to combine AI code analysis, full repository analysis, repository monitoring and notifications into one working project.
What we learned
We learned more about Flask, APIs, AI model integration, GitHub Webhooks, deployment and working with larger codebases.
What's next for IssueAgent
We would like to add automatic testing of suggested fixes, better repository monitoring and potentially allow IssueAgent to create pull requests with proposed fixes.
Log in or sign up for Devpost to join the conversation.