From Bug Report to Pull Request: An Autonomous Agent Pipeline for Production Issue Resolution
Roberto Milev (Navan), Uday Kanagala (Navan), Chris Cholette (Navan)
Architectural Patterns & Composition Engineering & Operations
Summary
An end-to-end autonomous agent (Sherlock) that traces production errors from Jira through New Relic to GitHub and opens a fix PR, resolving 41% of tickets in ~9 minutes vs. a 4.2-hour manual baseline.
Description
Software teams at scale spend hours tracing production issues across disconnected systems — from issue trackers, through observability platforms, to the responsible codebase — before they can begin fixing them. We present Sherlock, an end-to-end autonomous agent that resolves production issues by orchestrating across Jira (issue tracking), New Relic (observability), and GitHub (source code). Given a Jira ticket containing an error description or screenshot, the agent autonomously traces the error through production stack traces, identifies the responsible service and code, generates a fix, and opens a pull request. Built on Claude SDK running on AWS AgentCore runtime, Sherlock includes a human-in-the-loop mechanism for graceful degradation when information is insufficient. Evaluated on 100 production tickets, the system achieved full autonomous resolution (PR created) for 41% of tickets in an average of 9 minutes, compared to a 4.2-hour manual baseline.
