What will I learn?
Discuss the modules, practical projects, learning approach and upcoming course arrangements.
Develop the practical skills needed for an AI Engineer in Test role.
From AI fundamentals and no-code labs to golden datasets, Python evaluations, CI/CD and red teaming. Build a complete AI Agent Quality & Evaluation solution.
Speak with an AI expert before deciding whether this course is right for you.
Discuss the modules, practical projects, learning approach and upcoming course arrangements.
Get the current course fee and ask about payment options before committing.
Prepare to test, evaluate and improve AI agents using practical quality engineering, automation, evaluation frameworks, CI/CD, security and governance skills. Employment is not guaranteed.
The course is designed for QA professionals, test engineers, automation engineers and test managers. Discuss your background, Python readiness and any relevant eligibility questions with an expert.
Concepts, no-code labs, Python evaluations and CI/CD. Explore each module below.
Understand LLMs, prompting, RAG, MCP, tools and agent architecture.
Hands-on outcome: Trace a request from the user through the model, knowledge sources and tools; create an annotated architecture diagram.
Build agents with platforms such as n8n, connect tools and knowledge, and evaluate responses and tool usage.
Hands-on outcome: Build a no-code agent and run at least 10 test cases, recording outputs, pass/fail decisions and metric scores.
Explore correctness, relevance, helpfulness, groundedness and safety, plus reference-based and reference-free evaluation.
Hands-on outcome: Create an evaluation scorecard and compare good and poor responses using multiple quality dimensions.
Design human-reviewed reference cases, expand coverage with synthetic data, and manage tagging, deduplication and versioning.
Hands-on outcome: Create 10 curated golden cases and 20–30 AI-generated cases, then review and organise the dataset.
Combine assertions, LLM-as-a-Judge, live experiments, trajectory testing, red teaming and production feedback.
Hands-on outcome: Map checks across pull requests, nightly runs, pre-release testing and production.
Use DeepEval, LLMTestCase, GEval and tool-call metrics, with custom rubrics and evaluation reporting.
Hands-on outcome: Build a Python evaluation framework and run it against at least 10 agent outputs.
Connect golden datasets, pytest, quality gates, baseline comparisons and red-team stages to delivery pipelines.
Hands-on outcome: Create pull-request gates, nightly evaluations and pre-release red-team stages in a sample pipeline.
Explore privacy, bias, security, human oversight and audit evidence, with an overview of GDPR, the EU AI Act, UK regulation, NIST AI RMF and ISO/IEC 42001.
Hands-on outcome: Build a governance testing checklist, AI risk register and risk/test/evidence matrix.
Build or use an AI agent and create a complete quality and evaluation solution that brings together the course’s practical work.
Talk to an AI expertLeave your details and request a callback from an AI expert to discuss course fees, job opportunities, course content and eligibility.