Description

Introduction

As language models evolve from passive responders to active participants in complex workflows, a new frontier has emerged: Agentic AI powered by LLMs. This hands-on course enables learners to move beyond basic prompt engineering to construct fully autonomous agents capable of perceiving context, retaining memory, planning, and interacting with tools and environments. Through guided projects, participants will learn how to build intelligent systems that can reason, adapt, and act independently.

Prerequisites

Participants should have:

• Working knowledge of Python

• Basic familiarity with OpenAI or similar LLM APIs

• Experience using tools like LangChain, AutoGen, or Transformers (recommended but not mandatory)

• Understanding of core AI/ML concepts (helpful but not required)

Table of Contents

1. Getting Started with Agentic AI

• 1.1 What Is an Autonomous Agent?

• 1.2 Evolution of LLMs from Tools to Agents

• 1.3 Key Components of Agentic Systems

2. Core LLM Capabilities for Autonomy

• 2.1 Tool Use with LLMs

• 2.2 Reasoning with Chain-of-Thought and ReAct

• 2.3 Prompt Engineering for Autonomy and Delegation

3. Building Blocks of an LLM Agent

• 3.1 Goal Definition and Task Breakdown

• 3.2 Memory Management: Vector Stores and Episodic Recall

• 3.3 Planning: Looping, Branching, and Execution Control

4. Frameworks and Toolchains

• 4.1 LangChain: Agents, Chains, and Tools

• 4.2 OpenAI Function Calling & Assistant API

• 4.3 AutoGen for Multi-Agent Collaboration

• 4.4 HuggingFace Transformers + Toolformer

5. Guided Projects

• 5.1 Personal Research Assistant with Memory

• 5.2 Task Automator for Web & API Interactions

• 5.3 Multi-Agent Planning for Complex Workflows

6. Agent Evaluation and Debugging

• 6.1 Tracing and Visualizing Agent Behavior

• 6.2 Monitoring Output Quality and Fail Modes

• 6.3 Safe Looping and Guardrails

7. Real-World Use Cases and Deployment

• 7.1 Integrating with APIs, CRMs, and Data Pipelines

• 7.2 Agentic Apps for Knowledge Work and Support

• 7.3 Hosting & Scaling Autonomous Agents (e.g., on AWS, Docker)

8. Ethics and Risk Management

• 8.1 Preventing Overreach and Runaway Behaviors

• 8.2 Human-in-the-Loop Strategies

• 8.3 Transparency and Explainability in Agent Decisions

Autonomous LLM agents represent a leap toward more intelligent, responsive, and adaptive systems. This course gives learners the hands-on experience needed to construct, guide, and deploy these agents in real-world applications. With practical skills in design, implementation, and evaluation, you’ll be prepared to build the next generation of AI-powered tools and assistants.