Test Automation with AI

Module 1 – Prompt Engineering and Strategic Approach to LLMs

• Introduction: What is AI and Why is it Important for Test Automation?
  ○ The role of AI in testing processes and the paradigm shift it brings.
  ○ Introduction to the concept of “AI Orchestrator.”

• Fundamentals of Prompt Engineering
  ○ Definition: What is prompt engineering and why is it critical?
  ○ Six Principles of an Effective Prompt: Clarity, Context, Role Assignment, Structure, Tone, and Iteration.
  ○ Practical Exercise: Comparing poor vs. effective prompts.

• Introduction to Popular LLM Tools: ChatGPT and Gemini
  ○ Strengths of both models and their use cases in test automation.
  ○ “Hallucination” risk and the importance of validating results.

• Workshop: Generating Basic Test Assets
  ○ Producing functional test scenario ideas for a feature.
  ○ Creating simple (valid/invalid) test data.
  ○ Explaining an existing code block with AI.

Module 2 – AI-Assisted Coding with Cursor AI

• Introduction to Cursor AI
  ○ What is Cursor AI? Differences and advantages compared to VS Code.
  ○ Key features: AI Chat, Code Generation/Editing (Ctrl+K), and chatting with the codebase via @Codebase.

• Developing Test Scripts from Scratch
  ○ Writing Selenium/Playwright code for a simple UI test (e.g., login scenario) using natural language.
  ○ Creating classes and methods aligned with Page Object Model (POM) principles.

• Understanding, Improving, and Debugging Existing Code
  ○ Refactoring: Making existing tests more readable and efficient through AI guidance.
  ○ Documentation: Auto-generating comments (docstrings) for functions and classes.
  ○ Debugging: Analyzing error messages with AI to identify root causes and solutions.

• Workshop: Automating a Sample Test Scenario
  ○ Automating key functions of a selected webpage (e.g., search functionality) from start to finish using Cursor AI.

Module 3 – Advanced Cursor AI Techniques and Integrated Workshop

• Advanced Cursor AI Usage: Productivity-Oriented Techniques
  ○ Project-Wide Refactoring: Using @Codebase to smartly replace specific code (e.g., an outdated locator) across all files.
  ○ Step-by-Step Code Generation for Complex Tasks: Breaking down a user story or requirement into multiple steps and guiding Cursor to generate code iteratively.
  ○ Test Maintenance: Analyzing flaky tests and making them more robust with Cursor recommendations.

• Integrated Workshop: End-to-End BDD Scenario Development
  ○ Step 1: Creating a BDD Scenario: Generating test cases in Gherkin format (.feature file) from a user story.
  ○ Step 2: Writing Step Definitions: Generating automation code for each step of the .feature file using Cursor AI.
  ○ Step 3: Completing Required Classes and Methods: Building Page Object classes and helper methods with Cursor.
  ○ Step 4: Live Debugging: Using Cursor’s debugging and chat features to resolve errors during test execution.

• Future and Best Practices
  ○ Ethical considerations and data privacy in AI-driven test automation.
  ○ Evaluating and critically reviewing AI-generated code quality.
  ○ The importance of building a shared “Prompt Library” for the team.

• Q&A and Closing
  ○ Summary and addressing participants’ questions.
  ○ Suggested resources for post-training personal development.