Arvo AI offers Aurora, an open‑source AI‑powered incident management platform for Site Reliability Engineers. Aurora uses LangGraph‑orchestrated LLM agents to automatically investigate alerts by querying over 30 integrated monitoring and observability tools, executing sandboxed commands in Kubernetes pods, and searching a knowledge base, then generates a structured root‑cause analysis with remediation recommendations. The solution is self‑hostable, free under an Apache 2.0 license, and supports major cloud providers and popular SRE tools.
Funding
Funding not disclosed
Founders
Product
Problem
Site reliability engineers often spend significant time manually investigating alerts, querying multiple tools, and stitching together data to identify root causes, which delays remediation and increases operational overhead.
Solution
Aurora by Arvo AI is an open-source, AI‑powered incident management platform that automates the investigation workflow. Using LangGraph‑orchestrated LLM agents, Aurora responds to alerts by querying over 30 integrated monitoring and observability tools, executing sandboxed CLI commands in Kubernetes pods, and traversing infrastructure dependency graphs. It searches a knowledge base for similar past incidents and synthesizes findings into a structured root‑cause analysis with remediation recommendations. The system runs on Python back‑end and a Next.js front‑end, and can be self‑hosted without per‑seat or per‑incident fees, making advanced AI‑driven incident response accessible to any SRE team.
Target Audience
Primary users are Site Reliability Engineers and platform teams responsible for monitoring, incident response, and post‑mortem analysis in cloud‑native environments.
Features
- Integration with 22+ services including PagerDuty, Datadog, Grafana, Slack, GitHub, Confluence, Dynatrace, and all major cloud providers (AWS, Azure, GCP, OVH, Scaleway, Kubernetes)
- Autonomous LLM agents that dynamically select tools, execute sandboxed commands, and retrieve logs, metrics, and configuration data
- Knowledge‑base lookup for historical incidents to accelerate diagnosis and suggest proven remediation steps
- Generation of structured root‑cause analysis reports with actionable recommendations presented in a web dashboard
- Open‑source Apache 2.0 license with Helm chart for Kubernetes deployment and simple CLI setup via make commands
- Support for custom “Aurora Actions” defined in natural language to trigger background automations during or after incident resolution