About the event

We’re thrilled to guide you on an exciting journey to mastering the art of experimenting, customizing, and deploying Foundation Models—think Large Language Models, Vision-Language Models, and Audio Language Models—on Amazon SageMaker.

Agenda

In this workshop, you’ll discover how to:

• Search, Select, and Deploy Foundation Models directly from the Hugging Face Hub 
• Set up a prompt engineering playground on SageMaker Studio using tools like Streamlit or the MLflow Prompt Engineering UI 
• Customize Foundation Models such as Llama 3.1 8b using strategies like Quantized LoRA-based fine-tuning 
• Deploy customized Foundation Models seamlessly to SageMaker endpoints for inference, and even expand your workflow to effortlessly leverage multiple LoRA adapters on the fly 
• Evaluate and compare fine-tuned models against widely accepted benchmark metrics (e.g., Toxicity, Classification Accuracy, Semantic Robustness) 
• Set up a robust monitoring service for your hosted Foundation Models using WhyLabs and LangKit 

Who should attend?

The builders we are targeting are the traditional persona served by SageMaker, Data Scientists and ML Engineers.