Enterprise Serverless Adoption. An Experience Report
summary
Legacy JBoss monolith is replaced with an event-driven serverless backend (switched in 2019) built on AWS Lambda and Step Functions with a GraphQL/cache front layer and Fargate-hosted UI; migration completes in 10 months by ~30 product squads, producing about 100 serverless microservices that use on‑demand data migration and GitHub Actions pipelines. The system automatically scales to handle millions of clickstream events and survives Black Friday; teams enforce observability, service audits, threat modeling and account-level isolation so new services ship independently and roll forward safely.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 0:25Agenda
- 0:59Sheen Brisals
- 2:38Struggles
- 5:29Why Serverless
- 7:50Benefits of Serverless
- 11:05Making a Start
- 14:17Moving to Serverless
- 16:52Accept Change
- 18:56Domain Driven Design
- 21:23Set Piece Play
- 22:42Bounded Context
- 25:32Proven Patterns
- 27:50Architecture
- 36:02Features
- 39:43Conclusion
related talks
39mServerless in production, an experience report
01h 02mServerless in production, an experience report
01h 02mFrom Monolith to Serverless
45mPatterns & practices for developing enterprise-scale serverless applications
31mAccelerate business innovation with serverless applications
47mServerless in production, an experience report
39mBuilding Highly Scalable Microservices with Serverless
53mPatterns and practices for developing enterprise-scale applications with serverless