Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off
One annual plan covers every course and certificate on Coursera. 40% off for a limited time.
Get Full Access
Learn how to transform manual issue triage from hours to seconds using an AI-powered system in this 19-minute conference talk from DevConf.IN 2026. Discover how maintainers and engineering teams can eliminate the 40% of time typically spent on manual issue analysis by implementing an open-source solution that automatically analyzes GitHub issues, identifies root causes, suggests concrete code fixes, and detects duplicates using your actual codebase as context. Explore the technical architecture built with Repomix for codebase snapshots, Google Gemini 2.0 Flash for AI processing, and GitHub Actions for integration, while learning about security measures including prompt injection protection and duplicate detection using semantic similarity. See how this working system currently supports GitHub with Jira integration planned, functions like a skilled triage engineer by understanding issues in full codebase context, auto-labels issues with type and severity classifications, and provides file and line-specific code fix suggestions. Watch a live demonstration on a real GitHub project showing end-to-end automated triage, and gain practical insights into prompt engineering for code analysis, building secure AI systems, scaling from tool to service architecture, and integrating AI into existing developer workflows to reduce repetitive work and improve consistency.
