Over the last few years, AI has gone from a set of emerging technologies to an industry obsession. Testers are being pressed to test products with AI features, and to use AI in their work. But too much of that work is “slop.” Too much of it is fake testing.
Whether designed for classification or prediction, or for generating text or code, all forms of AI have this in common: they are black boxes, imbued with magical behaviors and properties, whose behavior is neither controlled, nor understood, nor otherwise known to be safe. A responsible approach requires us to know enough about them to be aware of the risks—which means we must know how to test them. When we use AI to help us test, we must be able to critically analyze the work that AI does for us.
Testing has traditionally been framed as formalized, procedurally structured test cases that check output against specific, prescribed results. We never thought that was a good way to think about testing, but it is even less useful for testing AI technologies whose output is non-deterministic by design. The Rapid Software Testing approach—including its approach to automation—is designed to address that problem by building the test process around skilled testers rather than a set of canned practices or templates.
The RST methodology is focused on agency and skill, which come together to make responsible testing possible. We put testers at the center of testing, even as testers make extensive use of tools. In the age of popular AI, this is more important than ever. In a manner of speaking, all serious users of AI must become testers, now.
In this class, we will challenge you to use assistive AI to help you perform testing. Therefore you will need access to some sort of GenAI tool such as ChatGPT, Gemini, Claude, etc. We will also be talking about the use of agentic coding tools in testing, so you might also find it useful to have access to a tool such as Claude Code, Cursor, or Copilot.
RST/AI is taught online in a 3-day (three 90-minute webinars per day) format.
- To teach you how to plan and administer a test strategy that makes effective use of AI and other kinds of automation.
- To teach you how to approach the testing of an AI-powered product.
- To introduce you to RST methodology if you have not yet been exposed to it.
- If you are a tester with a basic understanding of how to use tools like ChatGPT and who needs to test AI or wants to use AI in testing.
- For best results, we recommend that you first take the foundational RSTE class. But it is not a requirement.
- If you are an experienced tester, we will help you put words to the tacit skills you have gained over time and provide exercises that help you refine them. You will appreciate that RST is a practitioner-centered methodology– you are in control.
- If you are a technical tester, you will learn how your technical knowledge and ability to write code can supercharge the testing process.
- If you are a developer who does some testing, your deep knowledge of product internals is both a crucial resource and a potential liability. You’ll learn how to improve the intrinsic testability of the product and how to manage your biases.
- If you manage people who test, you have the power to steer them and create an environment to help them do their most effective work. You will learn what good testing looks like, how to judge the progress of testing, and how to set high, yet reasonable expectations for the testing process.
- If you are a domain expert who does some testing, we will help you apply your deep knowledge of how the product is used and who uses it. You’ll learn how to find better bugs during acceptance testing—and help others do it too.
- If you work with people who test, you will gain an appreciation for the challenges of testing, discover how you can support the testing process, and learn what good testing can do for you.
RST/AI is taught Socratically, with exercises, discussions and illustrations of the RST methodology. Class discussions and debate address students’ questions and specific needs. We all learn from the unique perspective that each student brings to the class. We’ll cover:
- What Rapid Software Testing is and how differs from other methodologies
- The general nature of AI and why it is the hardest thing to test
- What test strategy is, why it matters, and how to construct one
- How to think about automation in testing
- How testing with AI requires a manageable tempo and constant vigilance
- How testing AI requires highly probabilistic and exploratory methods
- How agentic coding allows testers to make powerful test tools like never before
This is an online class:
- Live. The class is always taught live, because the instructor to responds to student questions and challenges.
- Timing. It takes place over three days, via Zoom. We do three 90-minute webinars each day, with 30-minute breaks between them, covering 5.5 hours total time daily. These can be time shifted to better accommodate certain parts of the world. If necessary, the online class can be arranged over non-consecutive days.
- Recording. For public classes, we don’t allow recording. For a private class we allow recording for internal use only.
- Interactivity. Interaction is an important part of the class. So, in addition to allowing students to self-unmute to respond to the instructor, we also run a MatterMost chat server during the class. This facilitates conversation among students, as well as with the instructor (publicly or privately). It also helps us share documents, links, etc. We encourage students to activate their webcams, but that is not a requirement.
How Do I Take This Class?
You can check the schedule to see if there is an upcoming public class.
Or you can email me at james@satisfice.com to inquire.