Summary
Disclaimer: This summary has been generated by AI. It is experimental, and feedback is welcomed. Please reach out to info@qconlondon.com with any comments or concerns.
The presentation discusses strategies to continuously ship software efficiently and safely while maintaining trust in rapidly changing environments.
Key Highlights:
- Trust and Safety: The presentation emphasizes engineering systems with trust and verification. It explores automated guardrails to maintain best practices across Monzo's extensive microservices infrastructure.
- Automation and Agents: Patel addresses the role of agents in code writing, pull requests, and orchestration of migrations. Despite their efficiency, they introduce new challenges, such as subtle logic bugs and increased change rates.
- Developer Experience: Patel showcases Monzo’s structured onboarding and documentation for new engineers, which includes standardization of operations, code structure, and development tools, enhancing developer productivity and code consistency.
- Continuous Integration and Code Checks: The presentation highlights the importance of reliable CI checks that introspect code deeply to prevent bugs and errors. This includes the use of AI tools to automate and enhance code review processes.
- Extending Tools and Processes Beyond Engineering: Non-engineers are encouraged to use Monzo's tools, promoting a broader organizational process improvement. This involves making processes accessible and reducing friction for all users.
- Infrastructure and Scalability: Monzo’s infrastructure leverages technologies like Kubernetes and Go to ensure smooth and scalable operations. The emphasis is on maintaining simplicity and familiarity across the organization to reduce learning overhead for engineers.
The presentation encourages organizations to invest in robust engineering practices, continuous improvement in developer experience, and safe implementation of automation to balance rapid iteration with maintaining trust and reliability.
This is the end of the AI-generated content.
Abstract
Moving fast only works if your systems are engineered with trust and verification in mind. In this talk, we showcase how teams can scale to the point where shipping hundreds of changes a day becomes boring, routine, and safe. We will discuss investing in automated guardrails that make it easy to follow best practices across Monzo's 3000+ microservices and stay on the paved road. These include services that power all our payments infrastructure and our core banking ledger.
Then we push that same idea “beyond humans”. Agents can continuously write code, open pull requests, orchestrate migrations, and make changes. They work extremely well within these guardrails, but they also introduce new failure modes, resulting in subtle logic bugs, code that works but is hard to reason about, and a drastically increased rate of change. We'll dive deep into the additional investments we've made (and are considering) to make the feedback loop for agents as fast and as reliable as possible.
Key takeaways:
- How to ship constantly without fear: the platform and process and guardrails that enable hundreds of deployments safely
- Understand how we use our monorepo and broad corpus of microservices so that both humans and coding agents can maintain existing patterns and utilise the paved road
- How we think about building and evaluating trust in the changes made by both humans and agents
- Identifying the systems that will break first when writing code becomes an instantaneous activity. - the idea is to push the audience to see that bottlenecks will move elsewhere.
Speaker
Suhail Patel
Senior Staff Engineer @Monzo Leading the Platform and Data Functions, Previously @Citymapper
Suhail is a Staff Engineer at Monzo focused on building the Core Platform. His role involves building and maintaining Monzo's infrastructure which spans nearly two thousand microservices and leverages key infrastructure components like Kubernetes, Cassandra, Etcd and more. He focuses specifically in investigating deviant behaviour and ensuring services continue to work reliably in the face of a constantly shifting environment in the cloud.