Generative AI

Level: Intermediate Duration: 90 minutes Download PDF

Generative AI

Creating new content with large language models and AI systems.

Learning Outcomes

By completing this topic, you will:

  • Understand transformer architecture fundamentals
  • Apply effective prompt engineering techniques
  • Integrate LLMs into applications
  • Evaluate and improve generated outputs

Visual Guides

LLM Capabilities
LLM Capabilities
Prompt Engineering
Prompt Engineering
Scaling Laws
Scaling Laws

Prerequisites

  • Neural Networks concepts
  • Basic understanding of attention mechanisms
  • API usage experience

Key Concepts

Large Language Models

  • Transformers: Attention-based architecture
  • Pre-training: Learning from massive text corpora
  • Fine-tuning: Adapting to specific tasks

Prompt Engineering

Techniques for better outputs:

  • Zero-shot: Direct instructions
  • Few-shot: Include examples
  • Chain-of-thought: Step-by-step reasoning
  • System prompts: Set behavior and constraints

Practical Applications

  • Content generation and summarization
  • Code assistance and debugging
  • Document analysis and extraction
  • Creative ideation support

When to Use

Generative AI excels for:

  • Tasks requiring language understanding
  • Creative content generation
  • Rapid prototyping of ideas
  • Augmenting human capabilities

Avoid when:

  • Exact numerical precision required
  • Full verifiability needed
  • Domain expertise is critical

Common Pitfalls

  • Hallucinations (confident but wrong outputs)
  • Prompt injection vulnerabilities
  • Over-reliance without verification
  • Ignoring token limits and costs
  • Not testing edge cases

(c) Joerg Osterrieder 2025