Delivering Multi-Cloud: Redesigning Our Cloud Network and The Team Behind It

Multi-cloud adoption is not new. Many organisations have been running multi-cloud for years, yet the hardest challenges surface in the infrastructure layer, especially cloud networking, as technology, scale, and business demands continue to evolve. This talk is a field report from a Senior Engineering Manager at Atlassian on what it really took to redesign and scale a cloud network across multiple providers while dealing with finite IP space, limited team capacity, competing priorities and tight delivery timelines.

In this talk, I will dig into how we rethought network design, adopted AI assisted workflows in our delivery and operations, and redesigned the team structures and mindsets to run multi-cloud and specialised cloud platforms in parallel. As we scaled the network to support the multi-cloud strategy, we reduced engineers’ context switching by 40%, and improved collaboration and flow in a remote-first team.

This session is for tech leaders who need to evolve systems and teams at the same time. You’ll learn strategies to progressively evolve architecture without disruptive rewrites; how to use lightweight fitness functions (for example, cost-effective network designs, change lead time, and DevEx metrics such as deployment time and efficiency) to guide evolutionary architecture decisions; and how to design team structures with clear roles, pod-based models that unlock flow in a remote-first setting. We’ll also look at how to balance continuous delivery with continuous discovery to deliver ambitious roadmaps without burning people out.