Enterprises often rely on Terraform modules to enforce rules - but this approach often doesn’t scale in the real world. In this session, based on real-world lessons from a large enterprise, we’ll explore the shift towards Azure Policy as the foundation of governance. Learn why the migration happened, the challenges faced, and how it enabled faster, safer adoption of cloud services. Attendees will leave with a roadmap for modernising governance, avoiding common pitfalls, and striking the right balance between freedom and control.

Almost daily now do we hear about another big data breach with often millions of people having their data stolen. Many of these data breaches happen to organizations that have infrastructure deployed in the cloud. In the aviation industry each incident is cause for a thorough review after which commercial operators usually adapt their operations or adopt new ways of operating to ensure this particular incident does not happen again. They do it successfully. Let's see if we can do this as well. In this session we will investigate real security incidents that happened to cloud-based organizations and find out ways to ensure we can avoid them in the future. This session will be jampacked with demos giving attendees a handy to-do list to take away back to their teams to work on and improve their cloud security posture.

Serverless makes it easy to start small. Building a security service that has to process large workloads reliably, survive partial failures and remain financially predictable is a different problem entirely. In this session, we will look at the architecture behind a real-world cloud security SaaS built heavily on Azure Functions and Cosmos DB, and examine the decisions that mattered once the system moved beyond simple request-response workloads. We will explore practical challenges including long-running and fan-out workloads, Cosmos DB partitioning and RU/s consumption, retry and idempotency patterns, Microsoft API throttling, concurrency control, failure recovery and multi-region resilience. Just as importantly, we will look at the trade-offs. When should you increase throughput rather than optimise queries? How aggressively should Functions scale when the API you depend on is throttling you? When does retry logic improve resilience, and when does it simply create a very expensive feedback loop? Using real architectural examples and code patterns, this session will show how to design serverless workloads that remain resilient under pressure without allowing reliability or cloud cost to spiral out of control. Attendees will leave with practical patterns they can apply to Azure Functions, Cosmos DB and other event-driven SaaS or enterprise workloads.