ISTQB AI Testing Exam Prep — CT-AI & CT-GenAI Certification
Prepare for the ISTQB CT-AI v2.0 and CT-GenAI specialist exams with syllabus-mapped lessons, hands-on Python and local-LLM labs, timed mock drills and a portfolio-ready AI test strategy.
✅ This course is included in the Unlimited & Global plan →About this course
<p>This course prepares manual and automation testers in India and worldwide for the two ISTQB AI specialist certifications: <strong>Certified Tester AI Testing (CT-AI) v2.0</strong>, which covers testing AI-based and ML systems, and <strong>Certified Tester Testing with Generative AI (CT-GenAI)</strong>, which covers using LLMs for testing safely. Every lesson maps to the official learning objectives and K-levels, and ends with interview questions and an exam-style quiz.</p><p>You will not just read - you will calculate metrics in scikit-learn, write dataset constraint and metamorphic tests, red-team a local LLM, build prompt chains for test design, a mini RAG test assistant and a guarded agent, sit two timed mini-mocks and finish with a portfolio-ready test strategy for a fintech AI feature. All labs use free tools, public datasets and synthetic data.</p><p>AI testing skills are in demand at Indian IT services firms, product companies and GCCs, and in remote roles for US, UK and EU clients where the EU AI Act, GDPR and India's DPDP Act shape how AI is tested. <em>This is independent exam preparation and is not affiliated with or endorsed by ISTQB or any member board; always confirm exam details in the official syllabus on istqb.org.</em></p>
What you will be able to do
- Explain the exam format, K-levels and chapter weights of ISTQB CT-AI v2.0 and CT-GenAI v1.1 and follow a 6-week study plan
- Map AI system behaviour to ISO/IEC 25059 quality characteristics and write measurable acceptance criteria for ML models
- Calculate accuracy, precision, recall, specificity and F1 from confusion-matrix data and challenge misleading accuracy claims
- Apply input data and model testing - dataset constraints, bias, metamorphic, adversarial, drift, A/B and back-to-back testing
- Run exploratory testing and red teaming against an LLM-based system and report reproducible findings
- Apply prompt engineering (structured prompts, few-shot, chaining, meta prompting) to test analysis, design and regression tasks
- Identify hallucinations, reasoning errors, bias, privacy and prompt-injection risks and choose mitigations
- Design an LLM-powered test assistant with RAG and guarded agents, and plan GenAI adoption in a test team
Syllabus
-
▶1. Exam Blueprints, K-levels and Study Plan — Plus AI Fundamentals for Testers Decode both ISTQB AI exam formats, K-levels and chapter weights, build a 6-week plan and learn the AI vocabulary from CT-AI chapter 1.
-
🔒2. CT-AI: Quality Characteristics, ML Workflow and Data — Where AI Behaviour Comes From Map AI behaviours to ISO/IEC 25059 quality characteristics, write AI acceptance criteria, and test data preparation and dataset splits.
-
🔒3. CT-AI: ML Functional Performance Metrics and Neural Networks — scikit-learn Lab Calculate confusion-matrix metrics by hand and in scikit-learn, challenge accuracy claims, compare model-dataset combinations and build a perceptron.
-
🔒4. CT-AI: Testing AI-Based Systems — Oracles, GenAI Testing and Red Teaming Lab Handle locked vs adaptive systems, solve the test oracle problem, explore and red-team a local LLM, and back-to-back test two models.
-
🔒5. CT-AI: Input Data, Model and Deployment Testing — Metamorphic, Adversarial and Drift Lab Test ML input data (bias, constraints, labels) and models (metamorphic, adversarial, drift, overfitting), then plan safe deployment - CT-AI v2.0 chapters 5-7.
-
🔒6. CT-GenAI: LLM Fundamentals and Prompt Engineering for Test Tasks — Ollama Lab Learn LLM basics, structured prompts, few-shot, chaining and meta prompting, and apply them to test analysis, design and evaluation on saucedemo.com.
-
🔒7. CT-GenAI: Managing GenAI Risks — Hallucinations, Privacy, Security, Energy and Regulation Identify hallucinations, reasoning errors and bias, defend against data leakage and prompt injection, estimate energy use and know the key AI regulations and standards.
-
🔒8. CT-GenAI: LLM-Powered Test Infrastructure, RAG, Agents and Adoption Roadmap — Lab Explain and build a mini RAG test assistant and a guarded tool-using agent, compare RAG with fine-tuning and LLMOps, and plan GenAI adoption in a test organisation.
-
🔒9. Full Mock Exam Drill — Two Timed Mini-Mocks, Explained Answers and Exam Strategy Sit two timed mini-mocks (CT-AI and CT-GenAI), review explained answers, build a timed drill runner and lock in an exam-day strategy.
-
🔒10. Capstone Project — Personal Revision Kit and an End-to-End AI Feature Test Strategy Build your exam revision kit and a portfolio-ready test strategy for a fintech AI feature that applies both CT-AI and CT-GenAI with runnable evidence.
