Sustainable Tech: Weak Code Ownership and Microservices

#158 - Sustainable Engineering Lessons From Scaling Up Wise - Balazs Barna

Topic 4 of 7

Published Dec 18, 2023 · Clip 14:18–32:25 (18:08)

Balazs explains how regional teams work at Wise with core systems and regional applications, and describes weak code ownership where any team can contribute pull requests to central codebases reviewed by the owning team. He recounts Wise's architecture journey from a monolith with three-hour builds, to microservices, to small per-partner microservices, driven by incidents like incompatible library versions breaking integrations. He details how lead-time data (a week for merges on core teams vs. two hours for owned services) drove decisions, and how his team built an integration library abstracting authentication, state machines, Wise API communication, and a back-office UI—now used in 20+ integrations and saving roughly four weeks per integration.