We provide virtual course about AI in Software Testing (M-AIST). You’ll learn how to harness large language models (LLMs), such as ChatGPT and GitHub Copilot, to generate, analyze, and maintain test cases with greater speed and precision.
Course overview:
Through a progressive series of labs, you’ll explore real-world techniques for AI-assisted test creation, legacy code analysis, code coverage improvement, exploratory testing, synthetic data generation, and much more. You’ll also tackle the unique challenges of testing AI systems themselves, manage flaky tests, and integrate AI-generated tests into CI/CD pipelines. Ethical considerations and model limitations are addressed throughout to ensure responsible AI adoption. Master the future of quality assurance with AI-powered testing.
After completing this course you should be able to:
By the end of the course, you'll have built a fully AI-enhanced testing workflow, from test generation to reporting, and gained the skills to apply AI effectively and confidently in your software projects.
• Boost Testing Productivity
• Improve Test Coverage
• Master AI-Powered QA Tools
• Level-Up Prompt Engineering Skills
• Tackle Real-World Scenarios
• Streamline CI/CD Workflows
• Write Maintainable, Clean Test Code
• Generate Documentation Effortlessly
• Work Smarter with AI, Not Blindly
• Capstone Project for Real-World Readiness
Course content:
Module 1 - Foundations of AI in Testing:
• Introduction to AI in Software Testing
• Benefits and use cases of AI for QA
• Overview of AI tools: GitHub Copilot, ChatGPT, Applitools, Launchable, and more
• Understanding zero-shot and few-shot prompting
Module 2 - AI-Driven Test Case Generation:
• Writing effective prompts for test creation
• Generating unit and edge case tests using LLMs
• Prompt patterns and strategies for maximizing test relevance
• Evaluating and refining AI-generated test cases
Module 3 - AI-Assisted Code Coverage and Refactoring:
• Measuring code coverage (line, branch, function)
• Using AI to detect gaps in coverage
• Refactoring verbose or redundant tests
• Mutation testing overview
Module 4 - Testing Legacy Code with AI:
• Understanding undocumented code with LLMs
• Generating regression tests for legacy behavior
• Using AI to reverse-engineer and protect critical functionality
Module 5 - Exploratory and Edge Case Testing:
• Defining exploratory testing and its value
• Generating edge cases with AI (fuzzing, boundary tests)
• Handling complex or malformed input scenarios
Module 6 - Generating Synthetic Test Data:
• Creating structured and unstructured data using AI
• Valid vs. invalid input generation
• Risks: hallucinations, unrealistic data, format constraints
Module 7 - Detecting and Fixing Test Smells:
• Common anti-patterns in test code
• Using AI to clean up, rename, and restructure tests
• Improving maintainability and test intent clarity
Module 8 - Testing AI and Machine Learning Systems:
• Unique challenges in testing non-deterministic output
• Output validation via heuristics, type checks, and human-in-the-loop
• Designing robust, behavior-focused test cases
Module 9 - Test Maintenance and Flaky Tests:
• Identifying causes of flaky tests (async, timing, randomness)
• Diagnosing issues with AI analysis of logs and failures
• Stabilizing tests with mocks, retries, and dependency control
Module 10 - CI/CD Integration:
• Incorporating AI-generated tests into CI workflows
• Using GitHub Actions for automated test runs
• Reviewing and tagging AI-generated content
• Managing regression lifecycles and metrics
Module 11 - Documentation and Reporting with AI:
• Auto-generating test documentation and summaries
• Writing JSDoc-style comments and QA reports
• Using LLMs for stakeholder-friendly communication
Module 12 - Limitations, Ethics, and Trust:
• Understanding hallucinations, overconfidence, and logic gaps
• Mitigating risk with prompt design and human oversight
• Intellectual property and authorship concerns in AI-generated code
Module 13 - Capstone Project:
• Apply AI techniques to a full-stack JavaScript application
• Generate, refactor, document, and integrate tests
• Demonstrate your complete AI-enhanced testing workflow in CI/CD
Target audience:
This hands-on course introduces software testers, QA professionals, and developers to the practical use of artificial intelligence in modern testing workflows.
Prerequisites:
Basic JavaScript Knowledge:
• Understanding of functions, variables, conditionals, and arrays
• Ability to write and read simple JavaScript code
Familiarity with Node.js and npm:
• Able to install packages and run scripts from the command line
• Experience initializing and managing a Node.js project
Introductory Testing Experience:
• Understanding of what unit tests are and how they’re used
• Exposure to a JavaScript testing framework like Jest is helpful but not required
Comfort Using the Command Line:
• Navigating directories and running basic terminal commands
Basic Git/GitHub Skills (for CI/CD labs):
• Cloning a repo, committing changes, and pushing to GitHub
• Creating and modifying GitHub Actions workflows (optional but beneficial)
Access to Required Tools:
• A computer with macOS, Windows, or Linux
• Internet access and a modern web browser
• A code editor (such as Visual Studio Code)