Dr.Code-II

Inspiration

I built Dr.Code-v2 because code review often becomes a bottleneck. I wanted a self-hosted tool that could reduce repetitive manual pull request work, react automatically to GitHub events, and make the review loop faster and more useful.

What it does

Dr.Code-v2 is a self-hosted, AI-augmented code analysis platform. It listens for GitHub pull request webhook events, fetches changed files, runs analysis, and helps send review feedback back into the pull request workflow. I also built a GitHub Integration Settings UI so the token, webhook setup, and integration flow can be managed inside the product instead of through scattered manual steps.

How I built it

I built the backend with FastAPI and structured the project around a dark-mode React frontend and a backend integration flow for GitHub events. I focused on making the system practical: handling webhook events, managing settings cleanly, structuring the API carefully, and building around a workflow that supports real pull request analysis instead of just static demos.

Challenges I ran into

The hardest part was getting the integration flow to feel real and stable. I had to think through webhook verification, error handling, and secret management while also keeping the project organized and within scope. Another challenge was resisting the urge to overbuild and instead staying focused on the most meaningful features.

What I learned

I learned that building a useful developer product means balancing capability with reliability. I also learned that clean scope, strong testing habits, and a clear integration story matter just as much as the headline feature.

Accomplishments that I'm proud of

I'm proud that Dr.Code-v2 is more than a concept — it is a working self-hosted tool with a real GitHub PR workflow and a usable configuration experience. I'm also proud that I kept pushing it toward something practical, understandable, and demo-ready.

What's next for Dr.Code-v2

Next, I want to keep improving the fix workflow, make the setup and validation experience smoother, and continue hardening the project so it becomes easier to maintain and more production-ready.