SESSION 1
Session 1 — OpenAI SDK Fundamentals
Text, vision, image generation, and text-to-speech from scratch.
2.5–3 hours•11 hands-on exercises
What you'll have built by the end
- ✓ A working Python script that chats with GPT from the command line
- ✓ A vision tool that can describe or compare local images
- ✓ An image generator that saves DALL·E output to disk
- ✓ A text-to-speech tool that produces mp3 files
- ✓ A multimodal pipeline that turns one sentence into a scene-by-scene illustrated audiobook
Prerequisites — tick these off first
Completed Session 0 or at least the What is an LLM pagePython 3.10+ installedSession1 folder with .venv activated and requirements.txt installedOPENAI_API_KEY set in Session1/.envVS Code with the Python extension and the .venv interpreter selected
Learning arc
The 11 exercises split into 5 progressive groups. Each builds on the one before — do them in order the first time.
1. Text
H1–H3
2. Memory
H4–H5
3. Vision
H6–H7
4. Generation
H8–H9
5. Integration
H10–H11
Exercises
1. Text