All problems
AI
Semantic Code Search in Large Monorepos
Code: ANX-RP-007
Build a semantic code search engine that understands intent rather than keywords, enabling developers to find relevant code across millions of lines.
Why it matters
Large codebases are impossible to navigate with keyword search. Developers waste hours finding existing implementations that could be reused.
Research gap
Existing code search tools are either syntactic (AST-based) or use basic embedding similarity without understanding code semantics.
Research questions
- 01 How do we represent code semantics that capture both structure and behavior?
- 02 What embedding architecture works best for cross-language code search?
- 03 How do we handle code that evolves over time?
Suggested technologies
Required skills
Team and supervisor
2-3 researchers, supervised by Muhammad Qasim.
Suitable for: CS, SE, AI.
Expected deliverables
- Code search engine supporting 10+ languages
- Evaluation on CodeSearchNet benchmark
- VS Code extension prototype
Work on this problem
Apply to the lab and this problem becomes your supervised research project.