At MenaDevs, I evaluate AI-agent-generated code and software-engineering outputs against task requirements. I also review AI-written pull request descriptions and commit messages, then provide structured feedback and rewrites to improve their accuracy and completeness.
I built YU-Bot, an academic advising RAG backend with hybrid search and contextual windowing. My projects also include AuraC2, an ICPC-style contest platform with live SSE scoring and Judge0 integration, and CodeLens, a semantic code intelligence backend.
