Generative_AI_in_seconds__Making_sense_of_unstructured_data_across_industries.pdf

genai

1. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Welcome slide

2. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Generative AI in seconds: Making
sense of unstructured data across
industries
Pawan K Mall
Senior Sales Engineer
Snowflake

3. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Popular Enterprise Generative AI Use Cases
Text analytics &
generation
Improve customer experiences
with accurate summaries and
analysis of reviews, service tickets,
chats, etc.
Reduce manual errors and
streamline processes with
automated document
processing
Document
chatbots
Increase productivity of
customer facing teams with
chatbots for user guides,
FAQs or wikis
Democratize knowledge with
employee Q&A chatbots that
can quickly search multiple
documents
Conversational
data analytics
Empower every user to
uncover insights without
having to write code or
analyze charts.
& more …

4. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Snowflake for Gen AI & LLMs: Platform Overview
Snowflake Cortex: Models & SearchSnowpark Container Services
USE AI IN SECONDS
Streamlit
OSS LLMs
Use Marketplace
Snowflake-Governed Data
Serverless AI & LLM Functions
Document AI
APPS IN MINUTESFULLY CUSTOM IN HOURS
Snowflake CopilotUniversal Search
Partner LLMs
Bring Your OwnUse Complete, Embed, Vector Search, and more…
Custom OrchestrationCustom UI
Fine Tuning
Make It Your Own
Amazon
Bedrock

5. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Snowflake Cortex
Why Use It
Quickly and securely analyze your data and build AI applications
contextualized with your enterprise data
How To Use It
Access the power of Snowflake Cortex via serverless SQL / Python
functions or as part of LLM-powered experiences such as Document
AI and Snowflake Copilot
Use Cases
•Cost effective way to build single tasks, including translating,
summarizing and sentiment analysis
•Build more complex prompt engineering on foundational models
(i.e. Llama) in conjunction with vector search functions to build
chatbot capabilities
Llama 2
forecast
General-PurposeSpecialized
complete
Snowflake CopilotDocument AI
complete()
embed()
...
translate()
forecast()
...
Serverless Functions
A fully managed service that hosts and serves pre-loaded, on-demand AI models, LLMs and
vector functions accessible via SQL or Python - without having to manage the infrastructure.

6. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Easy To Use
Use AI in everyday analytics or
build LLM applications without
infrastructure management
Cost Effective
Compute optimized for inference
and search to run where your data
is secure and governed
Flexible
Access industry-leading AI models,
LLMs and vector search
functionality via SQL / Python
functions
Snowflake Cortex: Key Takeaways

7. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Snowflake Cortex: Specialized LLM Functions
Execute fast
summarization
tasks that don’t
require
customization
Easily detect
customer
sentiment (Positive,
Neutral, Negative)
for customer
analytics
Extract key
information from
unstructured data
in a Question
Answer format
Fast and
continually
improving
translations from
any supported
language
SUMMARIZESENTIMENTEXTRACT_ANSWERTRANSLATE
Tuned to be more cost effective & performant for specific tasks. No prompt engineering required.

8. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Snowflake Cortex: COMPLETE Function
LLM Inference
COMPLETE(
model: string,
messages: string
)
Snowflake Tables
What Is It
Serverless SQL / Python functions that run
inference on performant LLMs across every
model-size category
Why Use It
Use LLMs from Mistral AI, Llama and
Snowflake without moving data out of your
Snowflake account or having to manage
infrastructure.
How To Use It
Use SQL / Python function in
Worksheets/Notebooks or Streamlit app
ACCESS MODELS FROM
CategoryModelContext Window
LargeMistral-
Large
32k tokens
MediumMixtral-8x7b32k tokens
Llama 70B4k tokens
SmallMistral-7B32k tokens
Gemma-7B8k tokens

9. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Process Unstructured Data with Document AI
Provides a first party multi-modal LLM through a visual
interface or SQL API to extract content easily from
documents using natural language.
Handwritten
Text
Document:Images Of
Documents
Documents With
Infographics
Scanned
Documents
DOCUMENT AI
Built-in, no-code
application with a large
language model
●Ask questions in
natural language to
extract information
from documents,
such as claims
reports, investor
reports, and
prospectuses
●Use UI to create fine-
tuned LLMs specific
to their documents
Create an LLM
for Business
Users:

10. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Document AI: How It Works
DOCUMENT AI SIMULTANEOUSLY COMPREHENDS ALL ASPECTS OF DOCUMENTS:
SEMANTICS
Document type:Agreement of Sales
CONTEXT
Seller:Naomi Lynnea Burr
Buyer:Eric Drake, Tammy Drake
LAYOUT
Sales Price:448,000.00
GRAPHICS USING MULTI-MODAL LLM
Buyer’s Initials:Yes
Seller’s Initials:Yes

11. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Bring LLMs & GPUs to Your Data with Container
Services
Open Source
LLMs
Partner &
Proprietary LLMs
Container
Services
SNOWFLAKE GOVERNED
DATA
GPUs
Third-party
DataInference
Fine Tuning
Key Benefits
●Bring the right LLMs to
your data that meets your
specific business use case
needs
●Host a range of services,
including GPUs for model
training, full notebook
experiences (e.g. Jupyter,
VSCode), and vector
databases (e.g. Pinecone,
ChromaDB)

12. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Build a front-end UI for LLM applications with
Streamlit
Key Benefits
●Turn data and models into
apps with open source UI
library for Python
●Deploy to Snowflake’s
secure and reliable
infrastructure in one click
●Maintain consistent
governance of data,
models and other objects
in Snowflake

13. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Call Centre Analytics with
Snowflake Cortex LLM and
Snowpark Container Services
Demo

14. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Call Centre Analytics with Snowflake Cortex LLM and Snowpark
Container Services
●Agents Efficiency KPI
●Call transcript
summarization
●Call Diarisation
●Call Intent
●First Call Resolution
●Customer sentiment
analysis
Snowflake Data Cloud
Key Outcomes
Try out at: https://developers.snowflake.com/solution/call-center-analytics-with-snowflake-cortex-and-snowpark-container-services/

15. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Thank you!
Please complete the session
survey in the mobile app
Thank you

16. © 2024, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Build beyond
skillbuilder.aws
Redeem your free 7-day
trial of AWS Skill Builder