Xây dựng Công cụ Hệ thống GenAI & Đa Tác tử cho Kiểm thử Phần mềm

Welcome to my course Build GenAI & Multi-Agent Systems Tools for Software Testing

In this hands-on course, you’ll learn to harness the power of Generative AI, AI Agents, and Multi-Agent Systems to build real-world tools for software testing. Whether you’re a QA engineer, SDET, or developer aiming to level up your automation skills, this course equips you with practical techniques to bring AI-driven efficiency into your testing lifecycle.

Today, QA engineers are no longer limited to writing test cases and checking logs manually. With the rapid growth of LLMs (like ChatGPT, LLaMA, and Gemini) and frameworks like LangChain and AutoGen, you can now build autonomous test agents, automate log analysis, and even create collaborative multi-agent testing systems. This course gives you the tools, patterns, and hands-on skills to make that leap.

By the end of this course, you will be able to:

• Understand the core concepts behind GenAI, AI Agents, and Multi-Agent Systems

• Run powerful open-source LLMs locally using Ollama (no paid API needed)

• Use LangChain to build intelligent tools and agents for QA automation

• Create custom tools that read PDFs, parse logs, and generate test cases

• Store and query data using vector stores with embeddings

• Build a RAG-powered agent that analyzes logs using context retrieval

• Develop a Test Case Generator Agent from product requirements

• Use Playwright with agents to simulate web scraping and behavior testing

• Orchestrate multi-agent collaboration using AutoGen and AutoGen Studio

• Construct fully automated agents that read requirements and output test cases

• Design multi-agent QA systems that mimic real QA workflows with minimal human input

Why This Course is Unique

Most AI courses focus on chatbots or language tasks. This course goes deep into the testing lifecycle and shows you how to build intelligent, context-aware agents for software quality assurance. You’ll move beyond theory and actually build working tools that:

• Read your requirements

• Understand logs and test results

• Generate test scripts and summaries

• Work together as a team of AI testers

All using open-source tools, local models, and practical Python code.