Applications today don't exist in a vacuum. They typically communicate with other services, such as databases, message brokers, cache servers, and even AI models. These interconnectivity requirements make it hard to do local/inner-loop development and testing efficiently, often forcing developers to write lots of code and configuration that never gets to production.

And what about the surge of AI-powered applications? What happens when the applications start using advanced features, such as RAG? These applications integrate with various AI model providers (such as OpenAI, HuggingFace, and Claude), yet results from these models are non-deterministic. How do you effectively test these integrations without costing your organization lots of money?

What if there was a better way? What if you didn't need to write any extra code or configuration and things just worked out of the box, making your life simpler and easier?

We will show how you can easily and efficiently test these integrations, proving that testing AI-powered applications doesn't have to be hard!