Building with LLMs

What you'll learn

Prompting · 11 lessons, about 6 hours

1. What an LLM actually does FreeNext-token prediction, and why that produces useful behavior.
2. Tokens and the context windowHow text becomes tokens, and what fits in one request.
3. Prompt fundamentalsClear instructions, context and constraints: the 80% of prompting.
4. System prompts and rolesSet behavior once, separately from the conversation.
5. Examples in promptsShow the model what good looks like (few-shot prompting).
6. Structured outputGet reliable JSON back that your code can use.
7. Prompt chainingBreak big tasks into a pipeline of smaller model calls.
8. Choosing a modelTrade capability, speed and cost across model families.
9. Reasoning models and effortWhen models think before answering, and how to control it.
10. Streaming responsesShow output as it's generated instead of waiting.
11. Conversations and memoryAPIs are stateless: how chat history and memory really work.

Context & retrieval · 8 lessons, about 4 hours

1. Images and documentsSend screenshots, PDFs and charts as input.
2. EmbeddingsMeaning as a list of numbers, so text can be compared.
3. Similarity searchFind the most relevant text by comparing embeddings.
4. Chunking documentsSplit long documents so the right piece can be retrieved.
5. Vector databasesStore and search millions of embeddings quickly.
6. Retrieval-augmented generationAnswer from your own documents instead of the model's memory.
7. Making RAG goodHybrid search, reranking and citations.
8. Hallucinations and groundingWhy models make things up, and how to design against it.

Tools & agents · 6 lessons, about 3.5 hours

1. Tool useLet the model call your functions (function calling).
2. The agent loopModel → tool → result → model, until the task is done.
3. Model Context ProtocolA standard way to plug tools and data into any AI app.
4. Agent design patternsRouting, orchestrator-workers and when not to build an agent.
5. LLMs in automationsPut models inside scripts and workflow tools that run on their own.
6. Prompt injection and securityHow untrusted text hijacks models, and how to limit the damage.

Quality & production · 6 lessons, about 4 hours

1. EvalsMeasure output quality with test sets instead of vibes.
2. LLM-as-judgeGrade open-ended output with rubrics and a second model.
3. Cost and latencyPrompt caching, batching and routing to cheaper models.
4. Logging and monitoringSee what your LLM app actually did in production.
5. Shipping an LLM appDeploy, secure keys, rate-limit and handle failures.
6. Capstone: research agentA marketing research agent with retrieval, tools and evals.

Before you start

Lessons here build on an earlier phase:

• Phase 0: Tools

The placement test can mark off what you already know, so you only study what's new to you.

What you'll build

Research agent

A marketing research agent with retrieval, tools and evals. Read the tutorial.

• Combine retrieval, tools and an agent loop into one assistant
• Make every claim traceable to a source the agent actually looked at
• Grade the agent automatically with code checks on a small eval set

When you finish

Finish every lesson in this phase, or show you know it in the placement test, and you earn a certificate of completion, "Building with LLMs (Phase 1)", with its own link you can share.