Outline
In this workshop, we will cover the following topics:
- Using the OpenAI API (Application Programming Interface) to interact with LLMs
- Controlling the output of LLMs with parameter settings
- Generating structured output with the OpenAI API
We will work with the GPT-4o-mini and GPT-4o models from OpenAI. Note that we could also use local models, but this would require a bit more setup and is outside the scope of this workshop.
The focus of this workshop is on using the OpenAI API, so we will only discuss Python in passing. We will be working with Google Colab and optionally Visual Studio Code. If working with Visual Studio Code, we recommend that you install an AI coding assistant, such as Github Copilot. For beginners, this will help to get you started with Python.
The workshop is structured as follows:
| Topic | Content |
|---|---|
| Setup OpenAI | OpenAI Platform and the OpenAI API |
| Setup Google Colab | Setup Google Colab for coding |
| Exploring OpenAI | OpenAI models and various parameter settings |
| Exercise 1 | Prompting and parameter settings |
| API Tricks | Various tricks to using the OpenAI API |
| Exercise 2 | API Tricks |
| Structured output | Generating structured output with the OpenAI API |
| Exercise 3 | Structured output |