# Netflix Builds an Operational Knowledge Graph for AI Investigations

> Netflix is linking incidents, deployments and infrastructure in a shared data model to help engineers and AI agents investigate production problems.

Oossa · 2026-11-13 · https://oossa.com/en/netflix-builds-an-operational-knowledge-graph-for-ai-investigations

Netflix is developing an operational knowledge graph that connects applications, incidents, deployments, alerts and infrastructure dependencies. The company says engineers can spend about four hours investigating a typical application regression, often because they must piece together relationships from separate systems. Its graph gives human investigators and AI agents a shared model of those connections, and supports AutoRCA, Netflix’s internal system for root-cause analysis. Netflix says it aims to produce ranked, evidence-backed explanations; people will retain approval authority over proposed fixes.

## The facts

- Netflix’s correlation engine examines 13 types of operational signals.
- The graph is stored in QuipuDB, Netflix’s distributed RDF database.

## Why it matters

For Netflix engineers, connecting operational clues in one model could reduce manual investigation work, though the company has not said how much time the system currently saves.

## Sources & references

1. [Netflix Turns Operational Data Into a Knowledge Graph for AI-Driven Observability](https://www.infoq.com/news/2026/11/netflix-ai-driven-observability/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=AI%2C+ML+%26+Data+Engineering) – InfoQ, 2026-11-13

Last updated: 2026-11-13