The Past, the Present, and the Future of Platform Engineering
Talkat KubeCon + CloudNativeCon China 2025 · on Wednesday, 2025-06-11· at 14:30· Level 19 | Crystal Court I·
summary
Platform engineering on Kubernetes converges on a clear claim: build an API-first control plane that uses existing CNCF building blocks rather than reinventing core pieces. Examples: combine Crossplane controllers, Argo CD, Backstage and Knative scale-to-zero as blueprints; use Q and KCP for multicluster scheduling and tenancy; extend with agent layers like Dapper Agents that call LLMs via MCP. The hard problems called out are governance, observability, guardrails for agents, and translating tribal rules into machine-enforceable policies.
AI-written summary · about
chapters (16)
- 0:00Platform engineering history
- 1:02Developer to platform builder
- 2:05Containers and Kubernetes
- 4:34Platform patterns over time
- 6:00Modern platform architecture
- 8:50Complexity in platform teams
- 9:53K native and serverless
- 11:15Dapper and API-driven infra
- 13:20Custom APIs and controllers
- 15:26Platform blueprints and tools
- 17:56Future platform tools
- 18:36AI impact on platforms
- 20:48Multi-cluster challenges
- 22:13AI agents in platforms
- 24:38Community in platform eng
- 25:41Q&A
related talks
27mThe Future of Platform Engineering
33mThe Platform Engineering Journey
31mClosing panel: The future of platform engineering and what to build next
28mClosing panel: The future of platform engineering and what to build next
19mPlatform Engineering beyond the Buzzword?
46mBeyond DevOps: How and Why Companies Converge on Platform Models
30mPlatform as a Product- 28mClosing panel: The future of platform engineering and what to build next