From the course: Azure AI for Developers: Building AI Agents
Unlock this course with a free trial
Join today to access over 25,300 courses taught by industry experts.
Function calling - Azure AI Services Tutorial
From the course: Azure AI for Developers: Building AI Agents
Function calling
- [Instructor] Let's start with function calling. We will create a travel agent assistant that helps our travel agency answer customer queries. This notebook guides you through each step. We first need to create an environment file that contains the needed Azure OpenAI parameters, and we then load the Azure configurations in our notebook. We then define our functions. In our example, we will have three functions. The first function provides the countries that our travel agency services. The second function provides the travel budget for each country. The third function provides weather for each country. The first two functions, as you can see, does not require a parameter, while the weather function requires a country parameter. These are the JSON definitions of the functions. We then need to create the functions using Python. For now, we hardcoded the values, but imagine that you will call other external APIs to extract the needed values. The first and second function may call an API…
Contents
-
-
-
Understanding AI agents2m 57s
-
(Locked)
OpenAI API Assistants overview5m 21s
-
(Locked)
Function calling5m 34s
-
(Locked)
File search4m 56s
-
(Locked)
Code Interpreter3m 21s
-
(Locked)
Multi-agent collaboration7m 7s
-
(Locked)
Creating assistants in Azure AI Foundry2m 38s
-
(Locked)
Challenge: Create agents using Assistants API1m 54s
-
(Locked)
Solution: Create agents using Assistants API3m 43s
-
-
-
-