Part2: Agentic AIOps for Samsung Account Service — Automating RCA with a Multi-Agent System

Part 2 of a two-part series explaining how Samsung Account service applies GenAI to build an Agentic AIOps multi-agent system for automating root cause analysis.

The goal is to connect observability data into a single context to automate RCA and produce actionable remediation guidance, with the aim of structurally reducing MTTR.

Performance requirements include presenting root-cause candidates with supporting evidence within 5 minutes of anomaly detection; on 500 errors, automatically providing related services, suspected causes, and prioritized action guidance; and automatically distinguishing code/redeploy-induced issues from infrastructure problems.

All results must be shared within Slack-based real-time workflows to support collaboration.

Background: AIOps (term introduced by Gartner in 2016) historically relied on predefined rules and statistical models and struggles as microservices, multi-cloud, and container orchestration exponentially increase managed components; the AIOps market is forecast to grow at a 21.4% CAGR through 2032.