Hi Everyone,
Here we are, back again with yet another edition of Expert Talks in Chennai. We are thrilled to host you on 30th August 2025 at Awfis OMR, 111,Rajiv Gandhi Road, Old, Mahabalipuram Road, OMR, Kottivakkam, Chennai. Please join us from 10:30AM to 1:00PM.
Event details are as follows:
Agenda :
1. Welcome & Intros (10 mins)
2. Talk 1: AI Agents Demystified: Understanding the Pillars of Autonomous Intelligence (45 mins)
3. Break (15 mins)
4. Talk 2: Build an Task Manager AI Agent using LLM (45 mins)
5. Tea & Networking
Talk 1: AI Agents Demystified: Understanding the Pillars of Autonomous Intelligence
This foundational talk will solidify your understanding of AI agents from simple chatbots to sophisticated autonomous systems. We'll explore what makes an agent truly "agentic" - the ability to not just understand and generate content, but to take meaningful actions in the real world. We will learn about the types of agentic systems, understand the critical role of tools and RAG, and discover how modern protocols like MCP are standardizing agent communication. By the end of this session, we will have a clear understanding of AI Agents, tools & framework for deciding when and how to build agentic AI solutions for business specific use cases.
Outline
- Introduction: What Are Agents Anyway?
- The Types of Agentic Systems
- Tools: The Hands of AI Agents
- RAG and Agentic Knowledge Systems
- Modern Agent Protocols and Standards
Talk 2: Build an Task Manager AI Agent using LLM
In this talk, we will explore AI agents by building a fully functional "Conversational Task Manager" from the ground up. This talk will demystify the core components of an AI agent, by demonstrating their direct implementation using Python and local Large Language Models (LLMs). By the end of the session, you’ll have a solid understanding of how to design and build AI agents using LLMs for practical use cases.
Outline
- Introduction & Setting the Stage
- The problem statement - Task Manager.
- Local LLM Setup and Framework Selection
- Building the Core Agent Components
- Demonstration
