01
LLM Evaluation
Correctness, relevance, completeness, consistency and safety.
AI QA GUIDE
AI QA is the quality engineering discipline for AI-powered products.
AI systems need more than prompt checks. Quality has to cover the model, knowledge, agent behaviour, security, user experience, accessibility, automation and production lifecycle.
02
RAG & Grounding
Retrieval quality, evidence, hallucination, missing and stale knowledge.
03
AI Agents
Tools, arguments, orchestration, memory, recovery and handovers.
04
Security
Prompt injection, data leakage, authorisation and unsafe actions.
05
Conversation
Multi-turn context, intent, clarification, escalation and customer outcomes.
06
Accessibility
Web, mobile and conversational AI accessibility.
07
Automation
API, UI, Playwright and AI evaluation automation.
08
Regression
Versioned critical journeys and release evidence.
09
Production
Quality signals, failure discovery and continuous improvement.
APPLY THIS TO YOUR PRODUCT
Have an AI system that needs a quality strategy?
Use the guide to understand the layers, then work directly with Suresh to assess or engineer them.