In a distributed system, effective observability helps us to understand how our system behaves during normal operation, and to quickly detect degradation.

Unfortunately, the observability of an individual service is not often brought into the scope of automated testing.

This talk will show how, by formalising the behaviour we require from our services in order to export telemetry data to tools like Prometheus (for metrics) or Honeycomb (for distributed tracing), we can make use of automated tests to build confidence in the observability of our system.

The introduction of a new programming language is a significant undertaking with long-term implications — a choice not to be taken lightly.

TrueLayer's main business is not system programming, yet we decided to bet on Rust. Why? How?

During the talk we will walk through our adoption story: from the first CLIs and cheeky projects to a Rust-backed new product line, with Rust-specific job openings. We'll share our expectations, our challenges, our doubts, the mistakes we made and the lessons we learned as we ramped up our usage.

If you are looking to embark on a similar journey, even if for a different language, this talk is for you.

We have used Rust to build our core banking stack. At the heart of the core banking stack is our ledgers which have been built using event sourcing and CQRS principles. We have iterated over our ledgers refining how we have implemented these patterns. We would like to give you an overview of what we do in core banking and take you on a tour of how our ledgers have evolved. We believe Rust has been a great choice through these evolutions and in many cases provides benefits over the traditional languages used for business applications.