1. Introduction to AI & Generative AI

• Evolution of Artificial Intelligence
• Generative AI fundamentals
• Real-world use cases across industries
  

2. Foundations of Neural Networks & LLMs

• Neural networks explained conceptually
• Evolution of NLP → modern LLMs
• Tokens, embeddings, capabilities and limitations
  

3. Understanding LLM Behavior

• Next-token prediction and probability
• Creativity vs determinism in AI outputs
• Hallucinations and factual reliability
  

4. Prompt Engineering

• Structuring effective prompts
• Zero-shot and few-shot prompting
• Role-based prompting and reasoning patterns
  

5. Productivity with Generative AI

• Email writing, summaries, reports
• Research and documentation workflows
• Developer productivity and coding assistants
  

6. Diffusion Models

• Conceptual understanding of diffusion models
• Image generation mechanisms
• Strengths and limitations
  

7. Multimodal AI

• Text-to-image and text-to-video systems
• Image understanding and captioning
• Cross-modal prompting
  

8. Retrieval-Augmented Generation (RAG)

• Why LLMs need external knowledge
• Simple RAG architecture
• Document-based question answering
  

9. Evaluating AI Output & Responsible AI

• Detecting hallucinations and bias
• Quality evaluation and validation methods
• Safe and ethical AI usage
  

10. Deployment & AI Ecosystem

• API vs on-premise deployment
• Token usage and cost basics
• Model landscape: GPT, Claude, Gemini, LLaMA
• Explore India's LLM landscape, featuring models like BharatGPT, Krutrim, and Sarvam AI  and local cultural nuances.
• GenAI tools and frameworks overview
  

Hands-On LabsParticipants will complete guided exercises including:

• Prompt engineering experiments
• LLM behavior exploration
• Tokenization and context experiments
• Multimodal AI demonstrations
• Document-based Q&A using RAG
• AI output evaluation and validation
• Responsible AI usage scenarios