Various offerings of learnings on how to use the LLMOPs template 

• Quick start guide​ 

  How to setup and configure the LLMOps Template - the first stops shown here are creating a repository, installing local dependencies, connecting to your ai:attack account and setting everything up in VS Code for local environment. Overview of Prompt Flow and Azure ML studio will be provided as well. We will present you with an example of how to setup your first prompts and then perform an evaluation flow on your local setup. The same run will then be also executed through the CI/CD pipeline to finish the session.

• How to RAG with the LLMOps Template 

  This learning will provide a comprehensive overview of the RAG (Retrieval-Augmented Generation) flow, a powerful approach that combines a model's generative capabilities with an external retrieval system. Participants will learn how the RAG flow works, with its four key components - Lookup, Generate Prompt Context, Prompt Variants, and Answer question with context - working together to fetch relevant documents or information and use them to improve the accuracy and relevance of the generated responses.

• Using the Snowflake connector with the LLMOps Template 

  This learning will provide guidance on how to set up a connection to a Snowflake database in order to fetch data for use in a PromptFlow template.

• Intro to Azure ML Pipeline with the LLMOps Template 

  In this video we will go through the process of setting up an Azure Machine Learning (AML) pipeline. What it means to create and register an environment will be showcased, followed by configuring and creating a computing resource. Afterwards the default pipeline steps (data connections, data processing, etc.) provided by the LLMOps template will be explained and the process of adding additional pipeline steps will also be shown. Lastly, creating and running the AML is done.