Idea · Wed, 28 Oct, 2026 · 07:51 am UTC
Smart Incident Detection and Debugging
Debugging production issues often requires manually correlating logs, API errors, deployments, database behavior, and service metrics across multiple tools.
Proposed solution
Build a system that detects unusual error patterns, groups related failures, correlates them with recent deployments, and produces an incident summary with likely causes and recommended investigation steps.
Suggested technology / approach
OpenTelemetry, structured logs, error tracking, metrics dashboards, alerting webhooks, and an LLM-based investigation layer. How I'd contribute: Observability architecture, backend integrations, incident workflows, and proof of concept.
How to contribute
I'd like to build it myself.
Notes
Start in read-only mode using existing monitoring data. Recommendations should be verified by engineers before any production action.
Submitted by
Amit Patil
