No items found.

Modular AI Systems with Feature, Training, Inference Pipelines on Hopsworks - a LLMOps Tutorial

Name: Modular AI Systems with Feature, Training, Inference Pipelines on Hopsworks - a LLMOps Tutorial
Start: May 08, 2024EDT
End: May 08, 2024EDT

May 8, 2024

6:30 pm

EDT

Microsoft, New York

Learn about a unified architecture for Batch, Real-Time, and LLM AI Systems based around three independent ML pipelines.

In this tutorial, we will introduce a unified architecture for Batch, Real-Time, and LLM AI Systems based around three independent ML pipelines

A feature pipeline to create feature data,
A training pipeline to train your model,
And an inference pipeline to make predictions with new data on your trained model.

We will use this FTI architecture to walk through a tutorial of building a LLM system that uses RAG and function calling to access structured data and a model.

---