Consistency Catalyst: The story of Paddle's in house microservice toolkit
summary
Paddle consolidates a chaotic stack into a Go toolkit used by most services since mid‑2022, delivering a service template plus libraries like go-http-server, go-bdd and go-scope and a client/export pattern; a small Apex team (~3 core engineers) builds and maintains the toolkit, teams adopt it via GitHub repo templates and Roadie scorecards, OpenTelemetry rolls out across services in weeks, time to get new services production‑ready falls substantially while dependency upgrades and occasional bespoke escapes remain the main friction.
AI-written summary · about
chapters (23)
- 0:00Paddle and toolkit journey
- 1:05Paddle tech stack overview
- 3:54Diverse tech stack challenges
- 7:01Shift to consistency in 2020
- 10:52Consolidating infrastructure
- 12:15Launching Apex team
- 15:00Shared libraries for problems
- 17:48Reusable boilerplate
- 22:48Service template for projects
- 24:52Case study: Toolkit adoption
- 27:22Merging libraries
- 29:29Toolkit impact on culture
- 31:17Measuring adoption
- 34:23Trade-offs in central decisions
- 36:52Cost of internal toolkit
- 39:01Recommendations for teams
- 41:05Q&A: Protecting libraries
- 43:35Q&A: Teams leaving path
- 46:28Q&A: Versioning libraries
- 47:55Q&A: Standardizing tools
- 50:04Q&A: Open source consistency
- 53:32Q&A: Apex team structure
- 57:05Q&A: Toolkit success
related talks
18mA case for microservices
31mGo + Microservices = Go Kit
01h 01mGo Kit A Toolkit for Microservices
30mFrom Service To Platform: A Ranking System in Go
39mGo + Microservices = Go Kit
34mIntroducing PIP.Services, a new open-source toolkit for enterprise-level microservice creation
22mHappy Path to Go – Transitioning teams to Golang
25mMicroservices, Monorepos, and the Efficient Build Processes