Generative AI (GenAI) Course

Lesson 1 • The GenAI Ecosystem Today

Maps the current landscape of providers, open-source projects, and deployment platforms. Helps learners navigate tool choices relevant to their professional context.

Lesson 2 • History and Evolution of GenAI

Traces development from early statistical language models to modern foundation models. Provides context for understanding current capabilities and limitations.

Lesson 3 • What Generative AI Actually Is

Defines generative AI by contrasting it with discriminative and rule-based systems. Establishes precise vocabulary used throughout the course.

Lesson 1 • Tokenization and Text Representation

Explains how raw text is converted into tokens and numerical embeddings. Grounds later prompt-engineering decisions in an understanding of model input format.

Lesson 2 • Attention Mechanisms and Transformers

Demystifies self-attention and the transformer block without requiring calculus. Explains how context window size affects coherence and recall.

Lesson 3 • Inference, Sampling, and Output Control

Covers how models generate tokens at inference time and how sampling parameters shape outputs. Enables learners to tune temperature and other controls purposefully.

Lesson 4 • Pretraining and Fine-Tuning

Distinguishes pretraining on large corpora from task-specific fine-tuning. Clarifies what a model knows by default versus what requires additional training.

Lesson 1 • Chain-of-Thought and Reasoning Prompts

Introduces techniques that elicit step-by-step reasoning to improve accuracy on complex tasks. Connects reasoning prompts to measurable output quality gains.

Lesson 2 • Anatomy of an Effective Prompt

Breaks down the components of a well-formed prompt: instruction, context, input data, and output format. Establishes a repeatable structure for all subsequent prompting work.

Lesson 3 • Zero-Shot and Few-Shot Prompting

Contrasts prompting with no examples versus providing demonstrations. Teaches when each approach is appropriate and how to select high-quality examples.

Lesson 4 • Prompt Iteration and Testing

Establishes a systematic process for diagnosing prompt failures and iterating toward reliable outputs. Introduces lightweight evaluation methods applicable without formal ML tools.

Lesson 1 • No-Code and Low-Code GenAI Interfaces

Surveys playground environments, workflow automation tools, and embedded AI features in productivity software. Enables immediate application without writing code.

Lesson 2 • Cost, Latency, and Rate Management

Quantifies the cost and speed tradeoffs of different models and request configurations. Equips learners to make economically sound deployment decisions.

Lesson 3 • API Concepts for Non-Engineers

Explains REST APIs, authentication, and request-response cycles in accessible terms. Removes the technical barrier to experimenting with GenAI services directly.

Lesson 4 • Structuring API Requests for LLMs

Covers the system message, user message, and conversation history fields used in chat-completion APIs. Teaches how message structure affects model behavior.

Lesson 1 • Retrieval Strategies and Reranking

Compares dense retrieval, sparse retrieval, and hybrid approaches. Introduces reranking to improve the relevance of retrieved context before generation.

Lesson 2 • Embeddings and Vector Stores

Explains how text is converted to dense vectors and stored for similarity search. Provides practical guidance on selecting and configuring vector databases.

Lesson 3 • Evaluating and Improving RAG Pipelines

Introduces metrics for retrieval quality and answer faithfulness. Teaches iterative improvement of each pipeline component based on evaluation results.

Lesson 4 • Document Ingestion and Chunking

Covers strategies for splitting documents into retrievable chunks while preserving semantic coherence. Directly impacts retrieval precision and answer quality.

Lesson 1 • Preparing High-Quality Training Data

Covers dataset curation, formatting, and quality filtering for instruction-following fine-tuning. Data quality is the single largest determinant of fine-tuned model performance.

Lesson 2 • Training Configuration and Monitoring

Covers hyperparameter selection, learning rate scheduling, and loss monitoring during training runs. Equips learners to detect and correct training instability early.

Lesson 3 • Evaluating and Deploying Fine-Tuned Models

Establishes evaluation protocols comparing fine-tuned models against baselines. Covers model merging, quantization for deployment, and version management.

Lesson 4 • Parameter-Efficient Fine-Tuning Methods

Introduces LoRA, QLoRA, and adapter-based methods that reduce compute and memory requirements. Makes fine-tuning accessible without enterprise-scale GPU infrastructure.

Lesson 1 • Planning and Reasoning Strategies

Covers ReAct, plan-and-execute, and tree-of-thought planning patterns for agentic tasks. Connects planning strategy choice to task complexity and reliability requirements.

Lesson 2 • Reliability, Safety, and Human Oversight

Identifies failure modes unique to agentic systems including cascading errors and unintended actions. Establishes human-in-the-loop checkpoints and guardrail patterns.

Lesson 3 • Tool Use and Function Calling

Teaches how agents invoke external tools via function-calling APIs to extend their capabilities. Covers tool schema design and safe execution patterns.

Lesson 4 • Agent Architecture Fundamentals

Defines the agent loop: perceive, plan, act, observe, and repeat. Distinguishes single-agent from multi-agent architectures and their appropriate use cases.

Lesson 1 • Privacy, Data Protection, and Consent

Covers risks of personal data exposure through prompts, training data, and model outputs. Establishes data minimization and anonymization practices for responsible use.

Lesson 2 • Safety, Misuse, and Harm Prevention

Catalogs misuse vectors including disinformation, deepfakes, and automated manipulation. Teaches technical and policy controls that reduce harm potential.

Lesson 3 • AI Governance Frameworks and Compliance

Surveys emerging regulatory principles, voluntary standards, and internal governance structures. Equips learners to build compliant AI programs within their organizations.

Lesson 4 • Bias, Fairness, and Representation

Examines how training data and model design introduce bias into GenAI outputs. Teaches bias auditing methods and mitigation strategies applicable across modalities.