AI Testing Webinar: Beyond the Basics – A Practical Guide for Testers
In our first webinar, we learned to evaluate a single AI response. Now, we’re taking the crucial next step: building a holistic, repeatable testing strategy for an AI-powered agent.
How do you move from one-off manual checks to a robust framework that ensures quality from the code to the cloud?
This practical, hands-on webinar is the follow-up designed for testers who need a clear guide. Using a real-world Test Plan Agent as our case study, we will walk through a comprehensive 7-part strategy. You will see concrete examples for each step. And, you’ll learn how to blend your classic testing skills with new AI-specific techniques.
In this session, you will learn how to:
- Plan and Design a Modern Test Strategy: Create a holistic test strategy. It covers everything from the application’s internal logic to the AI’s output. And, you’ll design the core assets you’ll need, like “Golden Datasets.”
- Test the Service (Before Testing the AI): Apply your classic API testing skills to the service hosting the agent. Check for standard behaviors, like error handling and timeouts – before you even look at the AI’s output.
- Build an Sanity Safety Net: Write fast, automated “smoke tests”, checking the live AI’s output for content relevance, and performance benchmarks.
- Perform a Manual Quality Check: Use a “Golden Dataset” method. Establish a human-approved benchmark for the diversity of the AI’s content.
- Automate with an “AI Judge”: Teach a second AI model to act as your “Judge,” turning your quality checks into a powerful, automated regression suite.
The webinar is designed for manual QA and test automation experts, who are making their way into the new world of testing AI-based products.
Join me on October 29th, at 3pm CET / 10am EDT.
Register now to reserve your spot and receive the free “AI Test Strategy Starter Kit“!