Learn to access LLMs through Azure AI Foundry, manage costs and security, and integrate APIs into research workflows.
This hands-on workshop introduces researchers to using large language model (LLM) APIs through Azure AI Foundry. You'll learn the fundamentals of cloud-based AI services, including what an API is, how model deployments (instances) work, and the differences between authenticating with an API key versus signing in through a web interface. We’ll walk through creating and configuring model deployments, selecting appropriate models for different research tasks, and using the Azure AI Foundry interface to explore and test models.
Along the way, we'll cover practical considerations for working in the cloud, including network and data security, managing API keys, monitoring usage, and keeping projects within budget. By the end of the workshop, you'll understand how to access LLMs programmatically through Azure and have the foundation needed to incorporate cloud-hosted models into your own research workflows.
Prerequisites: Basic familiarity with coding in Python or other programming language (but API examples will be in Python). Participants will be given temporary access to an Azure account with permission to use Azure AI Foundry and deploy models.
Leticia Vega
Email