SUCHET MAHAMUNI@SuchetMahamuni

BugProof is an AI-powered developer debugging and verification platform designed to make software debugging faster, more reliable, and evidence-driven. Debugging often requires developers to manually search through unfamiliar code, reproduce failures, trace execution paths, identify root causes, develop fixes, write or update tests, and verify that the fix has not introduced regressions. BugProof brings these fragmented steps into a single workflow. A developer starts by submitting a bug report or failing test. BugProof investigates the target repository by searching relevant files, functions, references, dependencies, and tests. It then collects self-verified evidence from the actual codebase and execution environment. Using this evidence, the system develops a root-cause analysis and attempts to go beyond AI-generated hypotheses and claims facts that have been directly verified. BugProof can then generate a focused fix proposal, including the affected code and explanation. The fix is never applied automatically, the developer must explicitly approve it. Once approved, BugProof validates the patch, applies it safely, and runs relevant tests to verify the result. The complete investigation, evidence, root cause, proposed fix, approval, and verification results can then be presented as an evidence-based report. The key idea behind BugProof is that instead of treating an AI-generated answer as fact, BugProof separates what the AI suggests from what the system can actually verify through source-code inspection, command execution, patch validation, and testing

💻 Online Hackathon | ⏳ Build: 48 hours | 📅 September 25–27, 2026 The official IBM Bob 2.0 hackathon. Build what's next in AI-assisted development, alongside the developers and experts shaping it. ⭐ Get help from experts throughout the event 🤝 Join a team or build solo 💰 Prize pool: $12,000 🧑🏻‍💻 Register before the Kick-Off Stream to start building with Bob