Develop AI Agents and Multi-Agent System for QA Practice

Learn how to build ai agents and multi-agent systems with lm models, OpenAI and local llm options, and dynamic prompt templates that generate bdd test cases from Jira stories.

Learn to build simple sequential chains in LangChain that connect prompts to generate bdd test cases and produce a short summary report, enabling a basic multi-agent automation workflow.

Discover how LangChain tools and agents automate QA tasks by defining tools like open browser, navigate to URL, and click, then linking them with a chain agent.

Learn how memory enables continuous conversations with LangChain: use conversation buffer memory to store chat histories, or conversation summary memory to save tokens, powering intelligent AI agents.

Install the latest Python from the official site and verify with python --version, then create and activate a venv for your Jupyter notebook AI agents, installing the request package.

Install the Jupyter extension in Visual Studio Code, create notebooks, and run Python code interactively by selecting the appropriate virtual environment, installing LangChain OpenAI, to initialize an LLM.

Set up your OpenAI API key, store it in an env file named OpenAI API key, load it with dotenv, and initialize the LM to invoke a test query.

Set up a local llm with Olama by downloading, installing, and verifying the tool, pull a model from its library, and run it locally for data security.

Learn the foundations of prompt engineering for llm-based qa agents, mastering techniques like rag, chain of thought, few-shot, and react to generate precise, structured outputs, including bdd test cases.

Master chain of thought prompting and role assignment to create explainable, reliable QA test cases in BDD style, covering positive, negative, and edge cases.

Learn how react and few-shot prompting empower QA agents to reason with private knowledge and retrieve public resources to generate accurate BDD test cases and structured outputs.

Explore building a live prompt that generates BDD style test scenarios from user stories using lang chain and OpenAI, with role assignment, chain of thought, and few-shot techniques.

Explore retrieval augmented generation to empower QA agents with a company knowledge base, using chunking, embeddings, and a vector database retriever for contextually accurate responses.

Learn chunking and embeddings to convert Jira story descriptions into chunks, create embedding vectors, and store them in a vector database for rag-based llm retrieval.

Store numerical embeddings in a chroma vector database and retrieve the most relevant text chunks with a retriever to power AI agents for QA practice with domain-specific context.

Build a RAG QA workflow by retrieving relevant context from a vector store, formatting prompts, and using a GPT-4.1 chat model to generate BDD style test cases.

Build ai agents with human-in-the-loop to generate bdd style test cases by fetching Jira story details via rest api and extracting readable descriptions from Atlassian document format.

Prepare data for AI retrieval by chunking text into meaningful pieces—300 size with 50 overlap—using LangChain's recursive splitter and store the chunks in a Chroma vector database with embedding models.

Transform Jira story chunks into vector embeddings with a Nomic embedding model, store them in a Chroma vector database, and build a retriever to supply context for generating test cases.

Build a BDD test case generation agent using retrieval augmented generation on jira requirements. Use prompt engineering and GPT-4.1 to produce given–when–then bdd scenarios with a later human review.

This lecture adds a human in loop to an ai agent that generates bdd test cases with gpt four and rag, enabling qa reviewer edits and formatting before ci/cd deployment.

Explore running BDD test cases in a live browser with an autonomous AI agent using LangChain and WebDriver IO, translating natural language scenarios into browser actions.

Build a local WebDriver IO server with a Node.js express API to control Chrome via REST endpoints like open, navigate, click, and set value for Python Lang chain AI agent.

Define and assemble LangChain tools to drive a WebDriver IO server, enabling an AI agent to open a browser, navigate to URLs, and perform actions via API endpoints.

Build and run a WebDriver IO test automation AI agent that reads BDD scenarios from an Excel file, executes actions in Chrome, and generates ci cd ready test scripts.

Learn to build a multi-agent qa system with Landgraf, where autonomous agents and a supervisor coordinate converting Jira user stories into bdd test cases and automated tests.

Build a multi-agent system that generates and tests BDD cases from Jira user stories using embedding models, chroma database, and GPT-4.1 via Landgraff, coordinated by a supervisor QA manager.

Build a test automation agent that executes BDD steps in a browser via WebDriver IO, coordinating QA and automation agents in a LangChain multi-agent system to validate results.

Coordinate two agents—the Shiva agent and the Shiva test automation agent—via a supervisor agent, fetch Jira stories, generate BDD test cases, run automation, and report results.