AI Test Manager for AI Systems

Course overview

An AI Test Manager plans, controls, and evaluates the quality assurance of AI systems, from feasibility testing to ongoing operations. The focus is on test concepts for AI applications, the assessment of result quality, risks, and error patterns, as well as documentation for approval and audit. Special Features in Testing AI Systems AI systems provide probable rather than fixed results - classical testing logic is not sufficient for this. - Understand the differences between rule-based software and AI systems - Recognize typical error patterns: hallucinations, biases, instability with new inputs - Establish acceptance criteria for non-deterministic results - Clarify the role of the test manager in the AI project in relation to project management, specialist departments, and IT. Test Concept and Test Planning - Develop a test strategy for AI projects along the project phases - Define test cases, test data, and evaluation criteria - Establish quality criteria for language models and AI services - Set up test environments, tools, and responsibilities for repeatable tests - Plan effort, schedule, and resources for tests. Testing Language Models and Generative AI In generative AI, the quality of the responses determines the usefulness in everyday work. - Systematically build prompts and input variations as test cases - Evaluate responses based on reference examples and evaluation grids - Compare the behavior of different models and versions - Identify and document hallucinations and inappropriate content. Assessing Quality and Fairness - Systematically measure and compare result quality - Check for bias and fairness in results - Assess robustness against unexpected or erroneous inputs - Examine data quality and data foundations as prerequisites for reliable tests. Risks, Regulation, and Documentation Verifiable evidence is a prerequisite for responsible approval of AI systems. - Evaluate and prioritize risks for each application case - Translate requirements from the EU AI Act and data protection into test criteria - Document test results, decisions, and approvals in a traceable manner - Organize coordination with data protection, information security, and compliance. Monitoring AI in Ongoing Operations - Observe quality after implementation and recognize deviations - Translate feedback from operations into improvements - Manage the interplay between specialist departments, IT, and project management - Prepare reports on test results in a comprehensible manner for decision-makers. Upon completion, participants will be able to systematically plan, evaluate, and take responsibility for the quality assurance of AI systems.

Upcoming dates

• 2026-11-17T23:00:00.000Z · 2027-01-19T23:00:00.000Z · Virtuelles Klassenzimmer
• 2026-11-17T23:00:00.000Z · 2027-01-19T23:00:00.000Z · Aachen · Hybrid Learning
• 2026-11-17T23:00:00.000Z · 2027-01-19T23:00:00.000Z · Augsburg, Bayern · Hybrid Learning
• 2026-11-17T23:00:00.000Z · 2027-01-19T23:00:00.000Z · Berlin · Hybrid Learning
• 2026-11-17T23:00:00.000Z · 2027-01-19T23:00:00.000Z · Bielefeld · Hybrid Learning

More courses in this field

• Applied AI & Data Science for Advanced Learners

  Clarusway GmbH

• Machine Learning with TensorFlow

  Talentspring Academy Group

• AI and Machine Learning Specialist with AWS Certified Machine Learning Specialty

  New Horizons Mitte

• Learning to Prompt: Data Science and Analytics

  StackFuel GmbH