Loading video player...
Built an AI-powered SRE incident analysis agent using Claude Code CLI that auto-investigates PagerDuty alerts by orchestrating queries across Splunk, Datadog in a single workflow. Reduced incident RCA time from 30+ minutes to under 5 minutes by automating cross-validation of metric alerts against live Splunk error logs. Integrated PagerDuty via MCP, splunk and DataDog via an internal tool called Tarmac, which uses token based Rest API