
Examine generative AI architectures and the stack for integrating AI into apps. Learn to configure Python, Node.js, VSCode, and Cursor AI, build GPT-based helpers, and orchestrate agents for practical experimentation.
With 20 years in quality engineering, the instructor has led teams from 2 to 80, tested code for startups and banks, and shares insights across agile, cloud, and machine learning.
Understand what are the limitation, what you can and cannot do in all phases of a software testing project
Explore the anatomy of prompts for AI agents, detailing how to set context, audience, tone, problem, and technology, and how to specify output format for effective, tailored responses.
Explore how adding AI changes a traditional decoupled app architecture, introducing an AI layer with model APIs, vector databases, and orchestration to drive inference and agents.
Install python on windows via the microsoft store, which resolves dependencies automatically. Verify the install with python --version in the terminal and prepare the remaining dependencies for benchmarking.
Install the python pip utility, download and run get-pip.py, then use pip install to manage python-based dependencies like open ai for hands-on demos.
Install node.js and npm by downloading the prebuilt installer from node.org, verify with node -v and npm -v, and use npm install to resolve dependencies.
Install Visual Studio Code on Windows 11, configure Java tools, install Maven and the debugger, and add extensions like GitHub Actions to enable continuous testing and CI for Java projects.
Get a brief introduction on what are the main components of AI
How NLP actually makes the AI more human.
Understand what is machine learning and how algorithms make the core of AI
Understand the basics concepts around supervised Machine Learning
Gain basic understanding of Unsupervised ML and Clustering
In this lecture you will get a basic idea of how Reinforced Learning is working together with ML Algorithms
In this material you will understand how critical good quality training data actually is.
In this material you will understand how critical good quality training data actually is.
Link to android app -> https://play.google.com/store/apps/details?id=com.openai.chatgpt&hl=en&gl=US
Link to apple app -> https://apps.apple.com/us/app/chatgpt/id6448311069
Explore real-world ai use cases, including self-healing code, automated testing from production through software development, and a self-service chatbot, plus data generation, code adaptation, and ci/cd with chatgpt.
Learn how to build a self-enriching item workflow that uses prompts and foundation models to iteratively enrich, robustify, and finalize outputs, with reinforced learning for future improvements.
Explore two architectures for gen AI enrichment: a traditional manual prompting workflow and a largely automated custom-application approach with cascade prompts and API integration.
See how ChatGPT enriches a fintech user story by transforming a draft into a comprehensive, import-ready user story with acceptance criteria, functional and non-functional requirements.
Explore enriching content with an API client for a custom model, automating prompts, removing noise, and delivering focused results via post requests and token-aware workflows.
Explore a non-generative ai-based e-commerce recommendation engine built on artificial neural intelligence. Feed customer demographics and cart content into an unsupervised model to suggest items and refine with delta feedback.
Explore how to build an AI-assisted shopping recommendation engine using untrained data and GPT-4, demonstrating 'frequently bought together' suggestions from a Kaggle shopping cart dataset.
Explore ai assisted code coverage, or self healing code, where a model analyzes your code and coverage reports to generate unit tests for uncovered areas, reruns tests and reinforces learning.
Explore how to use generative AI to self-heal code within a GitHub workflow, automatically fix unit tests, improve code coverage, and retrigger continuous integration with human oversight.
Adopt analytics-driven, behavior-based testing for mass-market SaaS using BDD with Gherkin and NLP, mapping human-readable scenarios to given-when-then tests and user journeys.
Learn how a next-generation testing framework uses analytics-driven given-when-then acceptance tests, continuous in production monitoring, and trained models to prioritize scenarios and reveal new ones from user behavior.
Explore how generative ai and large language models power a self-service portal chatbot, bridging users to back-end it systems through apis and a trained model.
See how a trained model powering a self-service bot reads the washing machine user manual to answer troubleshooting questions and information on main programs and temperature by scanning the document.
Learn how tokens define text units in large language models, how text is decomposed into tokens, and how token counts affect prompts and api usage using tokenizer examples.
Discover how to access the ChatGPT API by managing billing and prepaid credit, adding a credit card, and generating an API key, since a subscription alone isn’t enough.
Generate a new secret key for your account in the API keys panel. Use it for demos and any get or post requests, and never share it.
Discover how to perform your first completion API call with curl, using the GPT-3.5 model, API key, and temperature settings while navigating tokenization and simple quickstart guidance.
Learn to call the DALL-E 3 image API with curl, supply an API key and prompt, set image count and size, and receive image URLs with Python and Node.js samples.
Explore AI agents as orchestrators coordinating prompts and large language models to solve tasks. Learn how they use workflow routing, prompt chaining, and feedback loops across APIs, databases, and code.
Watch AI agents generate, review, and refine a user story with acceptance criteria and readiness checks, then publish to Jira using an MCP protocol.
Demonstrate how AI agents define test cases from acceptance criteria, apply test design best practices, generate Jira-posted tests, and enable automated testing of a bank customer story.
Automate acceptance testing for an AI agent by simulating a flight booking from Boston to New York, using browser use as a ChatGPT operator and logging all steps and data.
Demonstrates Copilot-driven Playwright script generation to create automated tests from prior browser interactions, using the page object model and decoupled test data for new scenarios.
Learn how AI agents perform automatic peer reviews of pull requests by fetching the latest commit from a GitHub repository, generating review comments, and suggesting code improvements.
Explore shift left and shift right concepts in software development and computer science, illustrated through prompts, memory updates, and practical tips on test driven and behavior driven development.
Use ChatGPT to generate a high level ERP test strategy, with prompts at basic to advanced levels, canvas, and covering functional, integration, and performance testing with data integrity.
See how a test automation architect uses ChatGPT to design a framework for React and Node.js apps, selecting Cypress, Playwright, JMeter, Postman, and Azure DevOps.
ChatGPT is used to create and validate user stories for an ATM withdrawal. The lesson adds acceptance criteria and non-functional requirements such as data privacy and fraud prevention.
Generate user acceptance test cases from a story using ChatGPT, a template, and given–when–then scenarios to validate functional requirements.
Explore OpenAI Codex, a coding agent powered by Codex one from the O3 model with human feedback; connect to GitHub, code or ask, run, and configure with plus subscription.
Configure Codex by linking a GitHub repo, enabling multifactor authentication, and creating an environment with dependencies and environment variables. Set up internet access, install packages, and run Playwright-based automation.
Learn how to configure an agents.md file for AI agents in a Playwright automation repo, detailing test data, the page object model, config.json settings, and a complete pull request workflow.
Craft a proper agents.md with clear instructions, config, pages, tests, and usage notes, and ensure no tests run unless explicitly instructed.
Explore using OpenAI Codex to generate invalid and valid test data, expand data sets, and manage environment tasks and pull requests.
OpenAI Codex helps generate negative invalid test data to simulate failures, manage json test data, and review changes via the master branch and pull requests.
Explore what artificial intelligence agents are, and how orchestrators coordinate prompts, large language models, tools, databases, and workflows to automate complex tasks in software development.
Demonstrates an AI agent workflow to auto-generate acceptance tests, BDD scenarios, and Selenium tests from a local story using Flow Wise, enabling quick, secure payment scenarios for online shoppers.
Learn Flowise AI, a drag-and-drop engine to orchestrate AI agents and apps, building workflows with prompts, llm chain, sql queries, and js customization, running locally.
Install and run Flowise locally, configure Node.js and npm, and start the app on localhost:3000 to explore premade agents, a marketplace, and human-in-the-loop tooling.
Build an ai agent through prompt chaining, using a supervisor orchestrator and worker nodes connected to a ChatGPT wrapper, configured with an OpenAI API key and model.
Orchestrate two AI agents to read a story from a file, generate test cases and BDD scenarios, and write the results to disk using a read/write tool and prompt engineering.
Demonstrates using fire crawl to turn a website into language-ready data, then a retrieval augmented generation pipeline uses vector embeddings to generate playwright tests from a webpage.
Explore Microsoft Autogen, an open source framework for a generic artificial intelligence, featuring a no-code studio and multi-agent orchestration to build apps with Python or .NET.
Install Microsoft Autogen following official instructions in an isolated VM, create and activate a dedicated environment, install Autogen and OpenAI dependencies, and test a simple chatbot.
Define a chatbot agent with Autogen, supply a GPT API key from the environment, and explore built-in memory as you chat.
Learn how to autogenerate two talking agents that critique and refine a Python function, debating to reach a robust solution with unit tests, using autogen memory for memory handling.
Learn end-of-conversation criteria in AI agents using hard stops and goodbye triggers. See prompts append final solution when ready, with a factorial example.
This introductory course is designed specifically for Software Quality Engineers interested in leveraging the power of Generative AI to enhance their testing, automation, and productivity.
Throughout the course, participants will learn how to apply Generative AI techniques to automate the generation of test cases, simulate user behavior, create user stories, write better code, rapid start adoption of a new framework and generate test data. The course will provide a practical understanding of how AI can be used to improve software quality and boost productivity, rather than focusing on the underlying AI algorithms and models.
Topics:
Understanding Generative AI: An overview of Generative AI and its applications in software testing and quality assurance.
Automated Test Generation: Learn how to use Generative AI to automatically generate a variety of test cases, reducing manual effort and increasing test coverage.
API Access for Chat GPT and Google Bard: Understand how to call Bard and GPT with the help of customer made consumers
Automated User Story Creation: Understand how Generative AI can be used to generate user stories, helping to streamline the requirements gathering and software development process.
Test Data Creation: Learn how Generative AI can be used to generate test data, ensuring comprehensive and effective testing while saving time and effort.
Code Investigation: Discover how Generative AI can assist in investigating complex code, identifying potential issues, and understanding code dependencies.
Code Explanation: Learn how Generative AI can be used to explain code behavior, making it easier to understand complex code and improve code quality.
Boosting Productivity: Understand how these Generative AI techniques can be integrated into your existing workflows to boost productivity, improve software quality, and accelerate delivery times.
[Practical] See how to create a Performance testing Framework with CI/CD on cloud with AI
[Practical] Create an API Testing Framework with Java and RESTAssured with AI
[Practical] Code Quality Validation Framework for Java
See differences between CHATGPT and GOOGLE BARD
Google Cloud AI solution with model training - Vertex AI
AI-Powered Browser Automation – Master sophisticated AI operators for seamless, intelligent browser automation.
Self-Learning AI Agents – Develop adaptive AI agents capable of self-improvement, continuous evaluation, and autonomous decision-making.
Make agents communicate and solve issues with MSFT Autogen. An agent collaboration framework