To Succeed in Platform Engineering, Build Infrastructure Knowledge Graphs
summary
Argues that building an infrastructure knowledge graph — not just a list of services — makes implicit tribal knowledge explicit and raises both developer productivity and operational excellence. Recommends a flexible property-graph model that maps concrete entities (GitHub repos, CI pipelines, container registries, Kubernetes objects, AWS resources) and relationships, contrasts it with fixed schemas, and shows practical issues: Backstage as the main OSS catalog, identity mapping via Okta, incident/service links via PagerDuty, ETL sync frequency and cloud API cost.
AI-written summary · about
chapters (9)
- 0:00Platform engineering intro
- 3:10Operational excellence core
- 7:22Software catalogs as knowledge graphs
- 11:12Knowledge graphs for infrastructure
- 15:47Practical infrastructure knowledge graph
- 20:00Modeling catalog entities step-by-step
- 24:12Mapping infrastructure relationships
- 27:20Handling catalog properties
- 31:10Key takeaways
related talks
39mPlatform Engineering as a (Community) Service
29mKeynote: Building Bridges with Platforms: The Role of the Platform Engineer
15mThe Art of Platform Engineering
32mPlatform Engineering Is Not About Tech
41mPlatform Engineering for Developers, Architects & the Rest of Us
08mInterview of Horacio Gonzalez
38mPlatform Engineering in the Age of AI Teammates
14mContext Engineering for DevOps: Turn Your Infra into a Graph