Testing LLM-based Systems
Learn practical strategies to test and validate Large Language Model (LLM) systems. Discover how to ensure reliability, evaluate AI outputs, and maintain quality in real-world LLM-powered applications.
Read moreBlog - page 3
Explore older entries from the archive.
Test-Driven AI Development (TDAID)
Discover Test-Driven AI Development (TDAID) — a modern approach that merges Test-Driven Development (TDD) with AI-powered software engineering. Learn how to apply TDD principles to AI coding agents, build reliable feedback loops, and prevent regressions in non-deterministic systems. This guide explains why TDD is making a comeback in the AI era, how to structure agentic workflows around tests, and what practices help teams deliver high-quality, maintainable code with AI tools like Claude, Cursor, and Gemini.
Read moreUnderstanding Playwright Agents
A deep dive into Playwright Agents and the Model Context Protocol (MCP) — how Microsoft’s latest AI-powered Playwright release automates test planning, script generation, and self-healing browser tests across Chrome, Firefox, and WebKit.
Read moreAI + Chrome DevTools MCP: Trace, Analyse, Fix Performance
How DevTools MCP enables AI agents to record real performance traces (LCP/CLS/TBT), analyse them, and apply fixes—bringing Lighthouse-style audits into an iterative debugging session. Notes on INP (field) vs TBT (lab) included.
Read moreMermaid diagrams: When AI Meets Documentation
Learn how Mermaid diagrams improve technical documentation and AI workflows by expressing architecture as code that stays readable, maintainable, and current.
Read more