Front Matter
FM.3: Course Syllabi

Course A: Undergraduate Engineering

Prerequisites

Python programming (loops, functions, classes). Basic linear algebra (vectors, matrices, dot products). No prior ML experience required; Chapter 0 covers foundations.

Course A: Undergraduate Engineering

Focus: Foundations, using LLM APIs, building basic agents. Students leave able to build and deploy LLM-powered applications. This pathway spends the first five weeks on foundations because undergraduates typically lack exposure to attention mechanisms and tokenization; skipping this material would leave them unable to debug prompt failures or understand why a model generates unexpected output. The second half jumps to the applied chapters (APIs, RAG, agents) because the goal is practical competency.

14-Week Syllabus

WeekTopicsLab / Assignment
1ML and PyTorch FoundationsBuild and train an image classifier in PyTorch
2NLP and Text RepresentationBuild a TF-IDF search engine
3Tokenization and Subword ModelsTrain a BPE tokenizer from scratch
4Attention and Transformers (Ch 03 through 04)Implement scaled dot-product attention
5Decoding and Text GenerationCompare decoding strategies on GPT-2
6Working with LLM APIsBuild a multi-provider API client
7Prompt EngineeringPrompt optimization challenge (few-shot, CoT)
8Embeddings and Vector DatabasesBuild a semantic search system
9RAG FundamentalsBuild a document QA system with RAG
10Conversational AIBuild a multi-turn chatbot with memory
11AI Agents and Tool Use (Ch 22 through 23)Build an agent with MCP tool integration
12Evaluation and Observability (Ch 29 through 30)Evaluate an LLM system with automated metrics
13Production EngineeringDeploy an LLM application with monitoring
14Final project presentations; further reading: Emerging Architectures (Ch 34), AI and Society (Ch 35)End-to-end LLM application (team project)
Recommended Appendices

What Comes Next

Return to the Course Syllabi overview to explore other courses and reading tracks, or proceed to FM.4: How to Use This Book for a quick orientation on conventions and callout types.