๐Ÿงช
Quality Assurance & Test Engineers

AI for QA Engineers

AI changes how software is tested. Master it before it obsoletes your toolkit.

AI systems are non-deterministic โ€” they hallucinate, change with model updates, and break traditional test assertions. This track teaches QA engineers to test AI systems, use AI to generate test suites, and build evaluation frameworks that actually work.

Duration
6 weeks ยท 5hrs/week
Format
Live online + async labs with HYVE engineer code review
Prerequisite
Software testing experience; familiarity with one testing framework

Curriculum

01

How AI Systems Fail

5hrs

Hallucination types, failure modes, non-determinism, and why existing assertions break on AI outputs.

Lab: Document and categorise 20 real AI failure modes from a live banking chatbot.
02

LLM Evaluation Fundamentals

6hrs

BLEU, ROUGE, BERTScore, G-Eval, and custom rubrics โ€” when each applies and how to build an evaluation pipeline.

Lab: Build a pipeline scoring LLM outputs on accuracy, relevance, and safety.
03

Testing RAG Systems

7hrs

Retrieval quality, answer faithfulness, context relevance. RAGAS, DeepEval, and custom evaluation frameworks.

Lab: Full evaluation suite for a RAG system serving UAE property queries.
04

AI-Assisted Test Generation

5hrs

Generate test cases, edge cases, and test data at scale with AI. Validate and maintain AI-generated tests.

Lab: Generate 200+ edge case tests for a REST API using LLMs.
05

Regression Testing for AI Features

6hrs

Detect when model updates break your product. Snapshot testing, golden sets, CI/CD quality gates.

Lab: Build a regression suite detecting model degradation across 3 LLM versions.
06

AI Quality Monitoring in Production

5hrs

Real-time quality monitoring โ€” logging, sampling, feedback loops, anomaly detection, alerting.

Lab: Build a production quality monitoring dashboard for a live AI chatbot.

Learning Outcomes

  • โœ“Build LLM evaluation frameworks
  • โœ“Generate & maintain test cases with AI
  • โœ“Test RAG systems for accuracy and relevance
  • โœ“Create AI regression suites
  • โœ“Monitor AI quality in production

FAQs