Generative AI & Agentic AI Course in Hyderabad

Online weekend batches | 45+ hours | Beginner to advanced

Next batch starts 14 November.

Learn to build real AI applications with large language models, from your first prompt to multi-agent systems. You will work with LangChain, LangGraph, Google ADK, CrewAI, n8n and RAG, and build three full projects along the way.

  • 45+ hours of hands-on training, with coding in every module
  • No prior AI or machine learning experience needed
  • Basic Python is the main prerequisite
  • Three real-world projects: research agents, a customer service agent and a document Q&A system

Get the next batch dates and fee

Fill in your details and our course team will call you back.

What You Will Be Able To Do

By the end of the course

  • Build RAG applications that are ready for real use
  • Create multi-agent AI systems
  • Build stateful workflows with LangGraph
  • Build agents with Google ADK
  • Run teams of agents with CrewAI
  • Automate workflows with n8n
  • Test and monitor LLM applications
  • Deploy secure AI applications that scale
  • Work with both closed-source and open-source LLMs

Tools and technologies

  • LangChain and LCEL
  • LangGraph
  • Google Agent Development Kit (ADK)
  • CrewAI
  • n8n
  • OpenAI, Anthropic Claude and Google Gemini
  • Hugging Face, Ollama and vLLM
  • ChromaDB and FAISS
  • RAGAS and LangSmith
  • Model Context Protocol (MCP)
  • Streamlit, FastAPI, Pydantic and Python

Generative AI & Agentic AI Course Curriculum

17 modules plus bonus sessions. Every module has a hands-on coding exercise.

Module 1: Introduction to LLMs and Foundation (3.5 hours)

  • Introduction to large language models: GPT, Claude and Gemini, tokens and context windows
  • Setting up your development environment and managing API keys
  • Introduction to LangChain and LangChain Expression Language (LCEL)
  • Working with OpenAI, Anthropic Claude and Google Gemini in one application

Module 2: Prompting and LLM Parameters (2.5 hours)

  • Prompt engineering fundamentals and prompt templates
  • Few-shot, chain-of-thought and role-based prompting
  • Temperature, top_p, token limits, and tracking cost per call

Module 3: Output Parsing and Structured Responses (1.5 hours)

  • Pydantic, JSON and custom output parsers
  • Error handling, validation and retry logic

Module 4: Retrieval-Augmented Generation (RAG) (7 hours)

  • RAG architecture and where it is used
  • Embeddings and vector databases (ChromaDB, FAISS)
  • Text splitting and chunking
  • Building a complete RAG pipeline end to end
  • Re-ranking, multi-query and parent document retrieval
  • Corrective RAG (CRAG) with LangGraph

Module 5: Memory and Conversation Management (2.5 hours)

  • Conversation, summary and window memory in LangChain
  • SQL-based and custom memory storage
  • Handling long, multi-turn conversations

Module 6: Tool Calling and Function Calling (2.5 hours)

  • Defining tools and how tool calls run
  • Building custom tools, chaining tools and handling errors
  • Tool design patterns and security

Module 7: LangGraph, Stateful AI Workflows (5 hours)

  • StateGraph, nodes, edges and conditional routing
  • Sequential, conditional and looping workflows
  • Human-in-the-loop, checkpointing and persistence
  • Multi-agent systems and supervisor patterns with LangGraph

Module 8: Agentic AI Systems (7 hours)

  • AI agents and the ReAct (Reasoning + Acting) framework
  • Building ReAct agents with tools
  • Agent swarms, supervisor-agent and hierarchical systems
  • Sequential agent workflows and agent handoffs
  • Tracing and evaluating agents with LangSmith

Module 9: Google Agent Development Kit (ADK) (3 hours)

  • ADK architecture and when to use it
  • Building agents and multi-agent workflows with ADK
  • Custom tools and production deployment

Module 10: CrewAI Framework (4 hours)

  • Agents, tasks and crews
  • Sequential and hierarchical processes
  • Task dependencies, custom tools and collaboration patterns
  • Running a crew in production

Module 11: LLM Evaluation and Monitoring (4 hours)

  • Evaluation metrics for LLM applications
  • RAGAS for RAG and agent evaluation
  • LangSmith tracing, custom evaluators and monitoring

Module 12: AI Security and Guardrails (2.5 hours)

  • Prompt injection, data leakage and model risks
  • Input validation, output filtering, PII redaction and content moderation
  • Secure API usage and data privacy

Module 13: Model Context Protocol (MCP) (1.5 hours)

  • MCP architecture and use cases
  • Building your own MCP server
  • Using MCP inside agent systems

Module 14: Open Source LLMs (2.5 hours)

  • Hugging Face models and local deployment
  • Running models locally with Ollama
  • High-performance inference with vLLM

Module 15: Production Deployment and Best Practices (3.5 hours)

  • Application architecture and API design
  • Streamlit apps, including a PDF chat application
  • Caching, cost and latency optimisation
  • Containers and CI/CD for AI applications

Module 16: n8n Workflow Automation (1.5 hours)

  • n8n basics and your first workflow
  • Connecting LLM agents to n8n workflows

Module 17: Real-World Projects (5 hours)

  • Research agent system with report generation
  • Customer service agent with a knowledge base
  • Document Q&A system for company documents

Bonus Sessions

  • OpenClaw
  • Cursor and Antigravity for AI-assisted coding
  • SerpAPI web search for agents
  • Managing token cost and budgets

Projects You Will Build

Research Agent System

A multi-agent research workflow that collects information, writes a report and checks its own quality with LangSmith.

Customer Service Agent

A conversational agent connected to a knowledge base that handles multi-turn conversations.

Document Q&A System

An advanced RAG application that reads PDFs and answers questions about company documents, with evaluation and monitoring.

Prerequisites

  • Python programming, basic to intermediate level
  • A basic understanding of APIs
  • Some comfort with the command line
  • No prior AI or machine learning experience required

FAQs: Generative AI & Agentic AI Course

Do I need AI or machine learning experience?

No. The course starts with how large language models work. You need basic Python, a basic idea of how APIs work, and some comfort with the command line.

How long is the course?

The course is 45+ hours, from beginner to advanced level, and it includes three real-world projects.

When are the classes?

Classes are online, in weekend batches. The next batch starts on 14 November. Fill in the form on this page or call us to get the fee and the class timings.

What will I build during the course?

You build something in every module. The three main projects are a research agent system, a customer service agent and a document Q&A system.

Which AI models and tools will I use?

OpenAI, Anthropic Claude and Google Gemini, plus open-source models through Hugging Face, Ollama and vLLM. On the framework side you work with LangChain, LangGraph, Google ADK, CrewAI and n8n.

Talk to Our Course Team

Batch dates, fee and any questions about the course

Call: +91 91116 21112
Request A Call Back