Level 1 · Entry Foundations¶
Goal: get comfortable calling an LLM from Python, engineering prompts that work reliably, wiring in tools, and shipping a small working AI assistant.
This level is about building with LLMs — you'll treat the model as a powerful API and learn the engineering patterns around it. All examples use the Anthropic API (Claude) with Python; the concepts map 1:1 to OpenAI, Google, and other providers.
Modules¶
- The LLM Landscape & Setup
- API Fundamentals
- Prompt Engineering
- Structured Output
- Tool / Function Calling
- Conversation State & Memory
- Streaming
- Errors, Rate Limits & Cost
- A Simple Agent Loop
- Capstone — CLI Personal Assistant
What you'll need¶
- Python 3.10+ and a code editor.
- An Anthropic API key — created in module 1. Every code snippet that talks to the API needs it; snippets that are pure Python (parsing, loops, cost math) run without one.
- A few dollars of API credit. The whole level can be completed for well under $5 using the models suggested in the lessons.
By the end of this level you'll have built a CLI personal assistant with tools, streaming responses, conversation memory, cost tracking, and a small evaluation suite — the core skill set behind almost every LLM product.