Overview
AI, Data Science & Cloud Certificates from Google, IBM & Meta — 40% Off
One plan covers every Professional Certificate on Coursera. 40% off Coursera Plus Annual.
Unlock All Certificates
Learn how to transform traditional SRE incident response workflows by implementing secure, LLM-powered systems in this conference talk from SREcon25 Europe/Middle East/Africa. Discover how a large-scale SRE team successfully transitioned from fragmented tools and tribal knowledge to modern AI-driven workflows using the Model Context Protocol (MCP), a novel pattern for injecting real-time, local system data including logs, configurations, and tickets into LLM prompts while maintaining strict security controls. Explore the implementation of MCP paired with a domain-specific retrieval system built on OpenSearch and Bedrock Titan embeddings that enables semantic search capabilities across incidents, dashboards, and playbooks. Gain insights into practical strategies for driving adoption, safeguarding sensitive data, and achieving measurable reductions in incident resolution times through this real-world case study presented by Theofilos Papapanagiotou from Amazon.
Syllabus
SREcon25 Europe/Middle East/Africa - Modernizing Incident Response with LLMs, RAG, and the MCP
Taught by
USENIX