Get started with generative AI on Snowflake. You'll set up your environment, work in Snowflake Notebooks, and build a simple AI app that summarizes customer call transcripts as JSON and displays the results in Streamlit.

Navigating the Generative AI Revolution with Snowflake

What we'll cover in this course

How to successfully complete this course

Preparing your development environment

Download the Snowflake Notebook for the simple AI app

Build a simple AI app in Snowflake

Chapter 1 review - Question 1

Chapter 1 review - Question 2

Introduction to GenAI on Snowflake

Explore Snowflake Cortex AI functions for common AI tasks. You'll prompt models with AI_COMPLETE, use task-specific and helper functions, try newer functions for filtering, aggregating, redacting, embeddings, and image inputs, and bring it together in a Streamlit app.

Course Introduction

Setting up your environment

Note: Update to LLM Functions

Introduction to AI Functions and AI_COMPLETE

Snowflake Services Consumption Table

Task-specific LLM functions & Helper functions

Using the AI_COMPLETE function

Using task-specific Cortex LLM functions

Using Helper Functions

Additional Cortex AI Functions: Overview

Additional Cortex AI Functions: Hands-on Demo

Reading: AI_PARSE_DOCUMENT and AI_SUMMARIZE_AGG

Multimodal AI: Images as Inputs

Vector Embeddings with AI_EMBED and AI_SIMILARITY

Download source code for the Streamlit app

Getting ready to put it all together: Build a Streamlit app

Module recap

Chapter 2 review - Question 1

Chapter 2 review - Question 2

Chapter 2 review - Question 3

Chapter 2 review - Question 4

Chapter 2 review - Question 5

Chapter 2 review - Question 6

Snowflake Cortex's LLM-Based Functions

Customize LLM responses with Cortex Fine-Tuning. You'll learn how parameter- efficient fine-tuning keeps costs low, prepare training data, fine-tune Mistral-7b to respond in a specific style, test your model, and build a Streamlit app that drafts custom emails and text messages.

What is Cortex Fine-tuning?

Cross-Region Inference for Standard Models

Analyzing Customer Support Tickets

Preparing your Training Data

Starting the fine-tuning job

Inference using fine-tuned model

Streamlit application to auto-generate custom emails and text messages

Continue your Generative AI education

Chapter 3 review - Question 1

Chapter 3 review - Question 2

Chapter 3 review - Question 3

Chapter 3 review - Question 4

Chapter 3 review - Question 5

Chapter 3 review - Question 6

Chapter 3 review - Question 7

Chapter 3 review - Question 8

Chapter 3 review - Question 9

Customize LLM responses with Cortex Fine-Tuning

This course introduces learners to generative AI and how to implement common AI use cases using Snowflake. The course starts with an introduction to important concepts, the setup of the learner environment, and the building of a simple application. It’s followed by learning how to use Snowflake Cortex AI functions to accomplish many common AI tasks, from prompting and summarization to embeddings and image inputs, and ends with learning how to fine-tune foundation models to perform specific tasks. This course is for anyone looking to skill up on AI, but is particularly suited for data scientists, ML/AI engineers, and data analytics professionals. To be successful in this course, you should have a background in Python, GenAI, and LLMs.

Intro to Snowflake for Devs, Data Scientists, Data Engineers

Build generative AI apps on Snowflake with Cortex LLM functions, prompt engineering, and fine-tuning.


Generative AI Essentials with Snowflake

Instructor

Let's practice!

https://s3.amazonaws.com/assets.datacamp.com/production/course_55953/subtitles/course_55953_73bc9ae7e55f68d48f2b43768541e76b.vtt

Snowflake Cortex's LLM-Based Functions - Vector Embeddings with AI_EMBED and AI_SIMILARITY

Vector Embeddings with AI_EMBED and AI_SIMILARITY

Create Your Free Account