Master Advanced Agentic AI + LangGraph + RAG+Memory -AUG'26

Explore Python as a high-level, general-purpose language with an easy-to-read syntax, learn through practical, hands-on modules from foundations to advanced topics, and apply it across automation, web, data, and AI.

Learn how to install Python on Mac using Homebrew, verify versions, set up aliases for Python and pip, and test with a simple hello world in an interactive shell.

Install and set up PyCharm on your Mac, explore why an IDE helps Python development, and choose between brew and DMG installation methods for the community edition.

Learn to install PyCharm on Windows by navigating to JetBrains' official PyCharm download page, choosing the correct Windows exe or ARM64, and completing the installer.

Agentic AI represents an autonomous, goal-driven evolution of quality assurance, capable of writing test plans, generating test cases and scripts, and executing tests with memory, reasoning, tool integration, and action.

Explore how agentic AI addresses QA pain points like test authoring and regression triage, enabling autonomous log analysis, self-healing tests, and autonomous test case generation with Lankchain and Lankgraph.

Explore large language models (LLMs) as the core of agentic AI, learning how tokens and predictions enable code generation, summarization, and reasoning, plus paid vs open-source LLMs.

Learn how to select the right LLM for QA tasks by balancing paid and open-source options, context windows, pricing, and prompt engineering, while avoiding hallucinations and bias.

Master prompt engineering by crafting clear, specific prompts with context and tone to guide AI responses, test and refine prompts to maximize accuracy and reduce back-and-forth.

Explore seven common prompt types—instructional, contextual, formatting, open-ended, specific example, clarification, and comparative—and learn to choose the right prompt type for your use case to maximize AI results.

Explore a real-time demo of prompt types, including instructional, contextual, formatting, open-ended, specific example, clarification, and comparative prompts, with live examples for banking and eCommerce contexts.

Master prompt framework by applying zero-shot, few-shot, and chain-of-thought prompting; use reason plus act (react) and structured JSON outputs for effective AI prompts.

Install Ollama on macOS by downloading the DMG from olama.com and dragging it to the applications folder, or install via brew with brew install OLAMA.

Learn to run OLAMA on macOS, start the service via spotlight or terminal, verify it runs at localhost:11434, and use basic commands like start, stop, version, and help.

Learn how to install Ollama on Windows by downloading the official Windows exe, running the installer, and launching the Ollama UI after selecting the default or a custom install location.

Start Ollama on Windows from the start menu, open the web UI, and verify it runs at localhost:11434; or start the server via the command prompt with 'ollama start'.

Download OLAMA models and run prompts locally on macOS or Windows using OLAMA pull and OLAMA run, then manage models with OLAMA list.

Choose the best OLAMAS model for specific tasks by balancing reasoning, coding needs, and hardware constraints. Compare LLAMA 3.x, QWEN 3, DeepSeek coder, and Mistral models.

Understand the b parameter in olama models, how billion parameters store training knowledge, and how model size affects memory, reasoning, and coding performance, with practical recommendations for local setups.

learn to hit Ollama via curl, using the /api and /api/generate endpoints, pass a model and prompt, and format responses with stream or jq for readable JSON.

learn to use curl with the olama chat endpoint for multi-turn conversations, and compare it to the single-prompt generate endpoint.

Learn to generate an OpenAI API key, sign in to the API platform, assign it to a project, name and save the key securely, and use it for curl requests.

Save your OpenAI API key locally, set it in your shell or environment variable, and enable a small billing balance to use cheap models like 4.1 nano.

Learn to call the OpenAI chat completion API with curl, including setting the API key, choosing a model, and extracting content from JSON responses.

Explore the three roles in the OpenAI API—user, developer, and assistant—and learn how developer role sets persistent priorities, while user prompts drive queries and assistant references prior conversation.

Generate a Google Gemini API key by creating a project, save the key to your ZSHRC or environment, and verify it with an echo command before using curl.

Learn to connect to the Gemini API using curl with your Gemini API key, run REST requests, and extract the text from the response using jq.

Understand Lanchain as an open source framework for building LLM powered applications. Explore how it handles multi-step calls, prompt templates, memory, and external tool integration.

Install Langchain and five dependencies (Langchain core, OpenAI, Google Gen AI, and Langchain community for Olama) by editing requirements.txt, then verify with the test-lanchain-install script and pip list.

Integrate LangChain with the LLM client by adding getLangChainLLM, preserving the existing chat method, and enable provider-specific calls for OpenAI, Google Gemini, and Olama.

Integrate LangChain into a test case agent by introducing a prompts package and a test_case_prompts module, replacing hard-coded prompts with a structured LangChain prompt template and escaping JSON braces.

Integrate a test-case generating agent with lang chain, build a chain using the lang chain expression language, invoke the LLM, parse to JSON, and export raw JSON and CSV.

Compare lang chain and LangGraph; learn how LangGraph enables state management, conditional branching, multi-step workflows, retries, and human-in-the-loop approvals for robust llm agents.

Install and verify LanGraph by updating the requirement.txt in the repository, installing via pip, and validating with a smoke test; organize agents in a new src/agents_v3 package for LanGraph-based agents.

Explore LangGraph's core concepts: state, node, and edges, using typed dictionaries, partial updates, and graph construction with start and end nodes for sequential processing and accumulation.