AI and Incident Management: How to Reduce Manual Work and Speed up the Incident...
summary
ilert GmbH builds a SaaS incident platform that demos an OpenAI function calling conversational UI to produce structured on‑call schedules (bi‑weekly demo works but occasional hallucinations occur), a self‑hosted embeddings pipeline using an Alibaba model from Hugging Face plus a vector database (krand) to vectorize alerts and group hundreds of duplicates by cosine similarity with a tunable threshold, an integrated LLM writer that produces status updates and an “80% finished” postmortem in minutes, and opt‑out, JSON outputs and non‑destructive, auditable LLM telemetry.
AI-written summary · about
chapters (8)
- 0:00AI and incident management
- 1:52Incident response stages
- 4:18AI-assisted on-call scheduling
- 9:20Reducing alert noise
- 19:59AI for incident messages
- 23:48Automating postmortems
- 27:00AI integration principles
- 29:29Q&A
related talks
36mRisks and Opportunities of AI in Incident Management
30mAI Agents for Incident Management
26mBecoming AI engineers
26mAI & SRE: Embracing the Ultimate Frontier of Automation
34mCan automated responses to cyber incidents be supported by AI?
05mAI-Driven Incident Resolution – Hype or Reality
42mAI for security - friend or foe?
20mAutomated Incident Management