Agentic AI Workshop

Hands-on Agentic AI Workshop

The Agentic AI Workshop is a practical programme designed to introduce participants to the architecture, development and application of AI agents.

Participants explore how modern AI systems can move beyond simple question-and-answer interactions to systems that can reason through tasks, access knowledge, use tools, call APIs and execute multi-step workflows.

The workshop combines conceptual understanding with hands-on implementation using Python, LLMs, RAG, tools, APIs, agent frameworks and MCP-based integrations.

What is Agentic AI?

Traditional Generative AI applications typically respond to user prompts and generate content. Agentic AI extends this approach by combining language models with tools, knowledge, memory or state, decision-making and task execution workflows.

An AI agent can receive a goal, determine the steps required, access appropriate tools or information and work through a task according to the workflow designed by the developer.

Core Agentic AI Concepts

Agentic AI Architecture

Participants learn the major components of an Agentic AI system and how these components work together.

• User request or task
• LLM / reasoning component
• Instructions and agent behaviour
• Knowledge and RAG
• Tools and APIs
• Memory, context or state
• Agent orchestration
• Task execution and results

Tool Calling & API Integration

A key capability of Agentic AI is the ability to interact with external tools and services. The workshop introduces participants to the concepts behind tool calling, function calling and API integration.

Participants can build examples where an AI agent uses an external function or service to obtain information or perform a defined task.

RAG & Knowledge-Grounded Agents

Retrieval-Augmented Generation enables an AI application to retrieve relevant information from a knowledge source before generating a response.

Participants learn how RAG can be incorporated into Agentic AI workflows so that agents can work with institution-specific or application- specific knowledge.

• Document ingestion
• Knowledge retrieval
• Context construction
• LLM-based response generation
• Knowledge-grounded agent workflows

Model Context Protocol (MCP)

The workshop introduces the concept of Model Context Protocol (MCP) and its role in connecting AI applications with external tools and data sources through standardized interfaces.

Depending on the programme duration, participants can explore practical MCP concepts and examples of tool integration.

Agent Orchestration

Real-world AI applications often require more than a single model call. Agent orchestration helps coordinate multiple steps, tools and components within an AI workflow.

• Multi-step agent workflows
• State and context management
• Tool execution
• Workflow routing
• Human approval points
• Agent evaluation and monitoring concepts

Multi-Agent AI Workflows

Advanced workshop programmes can introduce multi-agent systems, where multiple specialized agents collaborate within a larger workflow.

Participants can explore concepts such as agent roles, coordination, task delegation and workflow orchestration.

Hands-on Agentic AI Development

The workshop is designed around practical implementation. Participants progressively build and test components of an Agentic AI application.

• Create an LLM-based application
• Build a basic AI agent
• Connect an agent to a tool
• Integrate external APIs
• Add knowledge using RAG
• Design multi-step workflows
• Explore MCP integration
• Test and evaluate the prototype

Agentic AI Workshop Learning Outcomes

By the end of the programme, participants can develop an understanding of:

• The evolution from Generative AI to Agentic AI.
• The architecture and components of AI agents.
• LLM-based reasoning and task workflows.
• Tool and function calling.
• API integration with AI agents.
• Retrieval-Augmented Generation.
• MCP and external tool integration.
• Agent orchestration and state.
• Multi-agent workflow concepts.
• Practical Agentic AI application development.

Agentic AI Use Cases

Tools & Technologies

Agentic AI Workshop Formats

Who Should Attend?

• Computer Engineering Faculty
• Information Technology Faculty
• AI & Machine Learning Faculty
• Engineering Students
• Computer Science Students
• Data Science Students
• Research Scholars
• Software Developers
• Technology Professionals

Prerequisites

Basic programming knowledge is recommended for the technical hands-on sessions.

• Basic Python knowledge is helpful.
• Basic programming concepts are recommended.
• Familiarity with APIs is useful but not mandatory.
• No prior Agentic AI experience is required.

Workshop Infrastructure

• Laptop or computer for each participant.
• Reliable internet connectivity.
• Projector or large display.
• Microphone and speaker / PA system.
• Suitable classroom or computer laboratory.

Agentic AI Workshops in Amravati, Nagpur & Vidarbha

NXTGEN Intelligence Academy conducts Agentic AI workshops for colleges, universities, faculty, students and organizations in Amravati, Nagpur and across Vidarbha. Customized institutional programmes can also be delivered across Maharashtra.