GENERATIVE AI SOLUTION SPOTLIGHT

Transforming financial services

with generative AI

In collaboration with

© 2024, Amazon Web Services, Inc. or its affiliates.

Today's hosts

Ray Strickland Oleg Blokhin

Sr. Partner Solution Architect AI/ML, Executive GM, Global Head Financial Services,
AWS Provectus

© 2024, Amazon Web Services, Inc. or its affiliates. 2

Agenda
• Introduction – AWS generative AI tools
• Amazon Bedrock
• Amazon SageMaker
• Better together: AWS and Provectus
• About Provectus
• Generative AI adoption journey
• Case study
• Solution demo
• Getting started with generative AI
• Q&A

© 2024, Amazon Web Services, Inc. or its affiliates. 3

AWS generative AI stack

Applications that leverage LLMs and other FMs

Amazon Q in Amazon Q in Amazon

Amazon Q Amazon QuickSight Amazon Connect CodeWhisperer

Tools to build with LLMs and other FMs

Amazon Bedrock Amazon SageMaker

Guardrails Agents Customization capabilities JumpStart Clarify Ground Truth

Infrastructure for FM training and inference

GPUs AWS Trainium AWS Inferentia Amazon SageMaker

Amazon EC2 AWS AWS

UltraClusters EFA
Capacity Blocks Nitro Neuron

© 2024, Amazon Web Services, Inc. or its affiliates. 4

 Choice of industry-leading
FMs from AWS Partners

Amazon Bedrock Customize FMs using

your organization’s data
The easiest way to build and
scale generative AI applications
with LLMs and other FMs Enterprise-grade security and privacy

© 2024, Amazon Web Services, Inc. or its affiliates. 5

 Automatic model fine-tuning
and distributed training

Flexible model deployment options

Amazon SageMaker
Build, train, and deploy Tools for ML operations
ML models at scale

Built-in features for responsible AI

© 2024, Amazon Web Services, Inc. or its affiliates. 6

Better together:
AWS and Provectus

Oleg Blokhin
Executive GM, Global Head Financial Services,
Provectus

© 2024, Amazon Web Services, Inc. or its affiliates.

Provectus overview
Mission: We bring the power of cloud, data, and AI to reimagine the way businesses operate,
compete, and deliver customer value.

Our company Competencies

• US, Canada, LATAM, EMEA • Generative AI
• 500+ employees • AI and machine learning
• AWS Premier Tier Partner • Data and analytics
• 90+ AWS certifications • DevOps
• Cloud migration

Open-source leadership Foundational accelerators

• Data QA: great expectations • Cloud-native data lake
• MLOps: Kubeflow project • MLOps infrastructure
• Open data discovery platform • Generative AI (text-to-text)
• UI for Apache Kafka

Industry practices AI solutions accelerators

• Financial services and insurance • Customer 360 personalization
• Healthcare and life sciences • Intelligent document processing
• CPG, retail, and ecommerce • Supply chain intelligence

© 2024, Amazon Web Services, Inc. or its affiliates. 8

Generative AI adoption journey
Off-the-shelf LLM enablement
Achieve productivity gains by equipping your knowledge workers with out-of-the-box LLM
applications, and establish a solid foundation for secure, compliant, fair, and responsible
AI adoption from day one.

Generative knowledge management

Extend out-of-the-box LLM capabilities with an enterprise knowledge base that includes,
but is not limited to, FAQs, internal policies, training materials, and information from IT,
HR, product, sales, and marketing departments. Empower knowledge workers to be
more productive in their day-to-day tasks.

Generative documents intelligence

Boost productivity for corporate roles dealing with large, unstructured documents
in underwriting, auditing, asset management, and legal affairs by reducing the time
required for quality document processing and understanding from days to minutes.

Generative business intelligence

Increase the productivity of knowledge workers by exposing them to corporate business
intelligence capabilities and KPIs through a natural language interface, to accelerate
critical business decisions.

Universal enterprise agents

All of the aforementioned features and components can be combined into a comprehensive,
easy-to-use product that serves as a universal generative AI companion for enterprise users.

© 2024, Amazon Web Services, Inc. or its affiliates. 9

Generative AI solution
for insurance underwriting

© 2024, Amazon Web Services, Inc. or its affiliates.

Generative AI solution for insurance underwriting

Solution overview Key features

The solution is a decision platform designed Streamlined document processing

to assist underwriters in processing large • Documents are summarized in minutes, instead
volumes of complex insurance documents. of hours or days of routine manual work.
It streamlines the distillation process by
High quality of summarization
extracting and summarizing information,
• Trained on the customer’s domain data, it
enabling underwriters to make decisions on
creates rich summaries on par with human
policy extensions, assess claim likelihoods,
SMEs.
and set appropriate premiums.
Flexibility and scalability of generative AI
• Distillation of structured and unstructured
documents in different formats, at scale.

© 2024, Amazon Web Services, Inc. or its affiliates. 11

Case study: transforming insurance
underwriting with generative AI

© 2024, Amazon Web Services, Inc. or its affiliates.

Case study: international specialty insurance company
Transforming insurance underwriting with generative AI

Challenge Solution

A specialty insurance and reinsurance company, Provectus developed a generative AI-powered

saw an opportunity to reinvent the underwriting solution to summarize text and extract
process with generative AI. Traditionally, the information from technical surveys. At the
client’s subject matter experts (SMEs) would core of the solution is Cohere’s Command
spend days manually reviewing 100+ page large language model (LLM) hosted on
technical surveys before making an underwriting Amazon Bedrock. Provectus adjusted the
decision. By utilizing generative AI, the survey in-context learning for the model to perform
report can be summarized in minutes, thus summarization and extraction tasks,
streamlining and scaling policy issuance, and specifically tailored to the engineering
increasing potential annual recurring revenue. reports in a PDF format.

© 2024, Amazon Web Services, Inc. or its affiliates. 13

Case study: international specialty insurance company
Transforming insurance underwriting with generative AI

Outcome Results

The client is now using generative AI in • Time to value reduced from 10 days
production to process and analyze surveys, down to 10 minutes.
generating reports that speed up their
• Complex technical reports analyzed are
underwriting team’s ability to make policy
judged to be as good as an SME ability
decisions. The solution reduces the
to highlight key aspects of the report.
undifferentiated heavy lifting required for their
underwriting team to manually create reports • Higher quality summarization as
that include an executive summary, positive and compared to SMEs.
negative points of interests, and cite relevant
KPIs. Today, the survey report can be
summarized in minutes, enabling the insurer to
make underwriting decisions faster, with less
human intervention.
© 2024, Amazon Web Services, Inc. or its affiliates. 14
How it works: Transit

architecture diagram
Gateway

VPC Monitoring

VPC

ECS Cluster

JumpCloud Amazon Private

Cognito Subnet
Documents
Metadata DB
Processing Pipelines

Generated
User UI Application API Gateway
Distillation
Orchestration Summaries
OAC
Service

UI App WAF Restrict

Data Pipeline
Source Code by origin Amazon
Tracing Monitoring
Bedrock

© 2024, Amazon Web Services, Inc. or its affiliates. 15

Solution demo:
survey report distillation

© 2024, Amazon Web Services, Inc. or its affiliates.

Solution demo: survey report distillation

© 2024, Amazon Web Services, Inc. or its affiliates. 17

Starting your
generative AI journey

© 2024, Amazon Web Services, Inc. or its affiliates.

Generative AI with Provectus and AWS
Discovery/Assess (1-2 weeks) Build/Measure (3-4 weeks)

• Data and infrastructure discovery • Generative AI prototype deployed to

• AI use case prioritization and adoption roadmap customer’s non-production AWS environment
• Expected business value • Summary of hypothesis results
• Success criteria • Reference architecture
• Complexity and risk estimation • The deployment plan to build the next
• Proposed candidate to build momentum phase as a feature rich, secure, scalable
solution in production
• High-level solution architecture based on
AWS and LLM services • Refined TCO

• Deployment plan • Ongoing maintenance requirements

• Total cost of ownership (TCO) estimates

Partner programs available to offset implementation costs

Check out the Provectus’ generative AI offering

© 2024, Amazon Web Services, Inc. or its affiliates. in AWS Marketplace 19
Questions and answers/next steps

• Please fill out the AWS event survey and let us

know what you thought.

• Find the Executive Guide attached and it will be

shared after the event, along with the presentation.

• If you have any further questions or wish to

schedule a call with AWS and Provectus experts, Provectus in AWS Marketplace
please contact [email protected] or About Provectus:
[email protected]. Provectus is an artificial intelligence (AI)
consultancy and solutions provider helping
businesses achieve their objectives through AI.

aws.amazon.com

© 2024, Amazon Web Services, Inc. or its affiliates. 20

Thank you!

In collaboration with

© 2024, Amazon Web Services, Inc. or its affiliates.