Let's call platforms what they really are, ways to provide something-as-a-Service
summary
Argues that platforms must be treated as internal platform-as-a-product and implemented as X‑as‑a‑service so devops principles scale, turning collaboration into consumable services rather than monolithic buys. Recommends buying commodity PaaS, building bespoke components when business needs demand, and shipping discoverable, self-service and managed offerings. Calls out Backstage for unified UX and Cobra for CLIs, warns bad developer experience leaks into poor platform UX and breaks adoption.
AI-written summary · about
chapters (14)
- 0:00Welcome and introduction
- 0:45DevOps and platforms connection
- 3:12Breaking silos in tech
- 5:41Full stack teams limitations
- 8:50Autonomy vs responsibility at scale
- 10:11Platform engineering purpose
- 13:22Platforms as a service model
- 15:48Internal platforms as products
- 17:14Challenges of internal services
- 20:03Treating users as customers
- 21:51First generation platforms
- 23:36Second generation platforms
- 25:47Developer experience in platforms
- 27:53Wrap-up and key takeaways
related talks
33mWhy You Should Treat Your Platform as a Product
27mPortals and Platforms, Two Ps in a Pod? How Good Interfaces Make for Good Operability
27mWhy Developer Platforms and a curated Dev Experience are a must have when scaling
18mSunrise of Platforms as Services (PaaS)
24mSignificance of Platforms in Product Development
44mPlatform as a Product: A Deep Dive into the Technical Foundations
47mBuilding software products distilled: customer value, engineering and delivery
24mFrom Kubernetes to PaaS to Developer Control Planes