Overview

Summary

These tutorials teach you the fundamental patterns of Giskard Checks through hands-on, step-by-step exercises. Follow along to build your understanding of the core concepts.

Available Tutorials

Your First Test Zero to first passing test in under 10 minutes — no API key or LLM required

Your First LLM Call Configure a generator and call a real language model — then use LLMJudge to evaluate the response

Multi-Turn Scenarios Learn how to test conversational flows and stateful interactions across multiple exchanges

Dynamic Scenarios Learn how to use callable inputs and outputs that read from the trace, making interactions context-aware

Test Suites Group multiple scenarios into a suite class and run them with a single await

Next steps

Once you can build scenarios, extend with your own validation logic: see Custom Checks to create domain-specific checks with FnCheck, Check subclasses, and more.

Your First Test has no prerequisites — no API key or LLM needed. Start there if you are new to Giskard Checks.

What You’ll Learn

Through these tutorials, you’ll learn how to:

• Design effective test cases for AI applications
• Handle both single-turn and multi-turn scenarios
• Use dynamic inputs and outputs for context-aware interactions
• Use LLM-as-a-judge for nuanced evaluation
• Group scenarios into suites and run them with a single await

Prerequisites

Before starting these tutorials, you should:

• Have completed the Install & Configure guide
• Have basic Python and async/await knowledge