What is Generative AI? Understanding the Foundation of Modern AI Agents #2

Subrata Kumar Das published a lesson on Generative AI as part of a new course, explaining the evolution from traditional rule-based software to modern AI agents. The lesson covers the shift from predictive AI to generative AI, the role of large language models, and the architecture of AI agents, with a practical example using Microsoft Foundry to build a dietician assistant.

Everyone is talking about AI Agents. But before you build an AI Agent, there is one concept you absolutely need to understand: Generative AI. Generative AI is the technology that transformed software from systems that simply follow rules into systems that can understand language, generate responses, reason through instructions, and assist users in a natural way. As part of my new course: I published the first lesson where we explore the journey from traditional software to Generative AI and understand why modern AI Agents became possible. 🎥 Watch the video here: Many developers jump directly into AI Agents, prompts, tools, and frameworks. However, without understanding the evolution of AI, it becomes difficult to understand: In this lesson, we start from first principles and build the foundation required for the rest of the course. For decades, software followed a simple pattern: Input → Rules → Output Developers explicitly defined every behavior. This worked well until humans started interacting with software using natural language. Imagine building a dietician chatbot. Users might ask: All of these questions are similar. Yet they are phrased differently. Supporting thousands of variations quickly becomes impossible with manually written rules. Machine Learning introduced a new approach. Instead of writing rules, we train models using data. Examples include: Predictive AI can make decisions. But it still cannot create content. A predictive model can answer: Fraud probability: 87% But can it explain why? Can it write a detailed report? Can it create a personalized recommendation? Not naturally. This limitation led to the rise of Generative AI. Generative AI creates new content. It can generate: Instead of selecting predefined responses, it dynamically creates new outputs based on user prompts. At the heart of Generative AI are Large Language Models. LLMs learn language patterns from enormous amounts of data and use those patterns to generate human-like responses. This is the technology behind modern AI systems such as ChatGPT, Microsoft Copilot, Gemini, Claude, and many others. Every Generative AI application follows a simple architecture: User Prompt → LLM → Generated Response Understanding this flow is critical because it becomes the foundation of AI Agent architectures. In the second half of the course, we will build an AI Agent using Microsoft Foundry. The architecture we'll implement is: User ↓ React + Vite Frontend ↓ Microsoft Foundry Agent ├── Instructions ├── Generative AI Model └── Web Search Tool ↓ Response Understanding Generative AI is the first step toward understanding this architecture. Throughout the course, we will build: A practical AI-powered dietician assistant that can: By the end of the course, you'll have a complete working AI Agent built using Microsoft Foundry. 00:00 Introduction 01:30 The Problem With Traditional Software 02:30 Why Rule-Based Systems Break 03:51 The Rise of Predictive AI 05:06 Prediction vs Creation 05:55 What is Generative AI 06:43 What is LLM? 07:36 The Generative AI Flow 08:18 AI Agent Architecture 09:06 Subra AI Dietician In the next lesson, we'll answer a very important question: If Generative AI can already answer questions, why do we need AI Agents? We'll explore: and prepare for building our first Microsoft Foundry Agent. If you're interested in AI Engineering, Microsoft Foundry, Azure AI, Agentic AI, or building practical AI applications, this series is designed for you. Happy learning — Subrata Kumar Das 🌐 subraatakumar.com