Scribd
Upload
80 views4 pages

Bedrock Doc 1

Uploaded by

Likkitha
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
80 views4 pages

Bedrock Doc 1

Uploaded by

Likkitha
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

AMAZON BEDROCK-BASIC UNDERSTANDING AND SOME KEYPOINTS

OBJECTIVE:
Introduced by AWS to use foundational models with ML and AI Applications and provides
tools for developers to build and scale generative AI applications using already pretrained
models.

KEY POINTS TO BE CONSIDERED:


 FOUNDATIONAL MODELS:
Bedrock offers access to FM models from the AI
startups such as
AI21 – A startup that aims to build AI systems for better understanding and respond
like the way human responds to the text which is basically of NLP technologies..
(Contextual understanding)

ANTHROPIC -An AI based company for safety and research focused on creating AI
models that are robust and assists the risks associated with AI technologies.

STABILITY AI -An AI based company which develops AI systems for better


enhancement of human creativity, productivity and better development and
deployment of the applications

 CUSTOMIZATION:
Bedrock allows the users fine tune their FMs using their own data.
Business can be able to adapt the models for better fitting in the use cases, improvement
in performance .For example text and image generation models can be used as a
example for such customisation

 API INTEGRATION:
Bedrock provides API to process the embedding of AI into web apps, or
other software solutions.It also provides operations such as Generating tasks based on
the prompt, Summarizing the text ,answering relevant informations and if not specified
the model must also able to answer the question,if the text is not being trained through
the fine tuning model,it should be able to answer like ,”Sorry, I don’t have access to the
above information like that.
 SCALABILITY AND RELIABILITY:
Leverages the robust infra of AWS. One of the main service is it can be able to
provide high reliability and one of the main trust that the applications can work
seamlessly

 MANAGED INFRASTRUCTURE:
Since it is a fully managed service,it can able to handle the infrastructure
management completely.This keypoint provides the developers to optimize their
applications without any worry about underlying infrastructure.

 SECURITY AND COMPILANCE:


Since we know that AWS is well known for its high security and privacy purposes,
likewise BEDROCK is well known for the protection of customer’s data which includes
Data Encryption, Access Controls, Compilance to meet with the users requirements.

 COST MANAGEMENT:
Same as AWS ,pay as you use model, highly manageable interms of cost and
the business can be able to scale their usage based on the upfront investments.

USE CASES OF BEDROCKS:


Bedrocks are used for various purposes in various industries and
appliactions such as
1)Content creation
2)Customer Support
3)Data Analysis and summarization
4)Creative designs for image generation models and so on…..
DIFFERENCE BETWEEN BEDROCK AND LANGCHAIN:

1) Langchain develops framework for creating the LLM models ,whereas Bedrock
provides scalability and access to scale the foundational models in a simple and
effective manner
2) Langchain provides tools tor prompt,deep understanding of the models and the
workflows but Bedrock simplifies the process of integrating AI by its
infrastructure and APIs
3) Langchain provides flexible access to various model of choice but in case of
bedrock it provides only for foundational model
4) LLMs requires various expertise for understanding and setup of the model but in
case of bedrock it is service based managed approach for various range of users
5) LLMs mainly focuses on developers and researchers,while the bedrock focuses on
the business for better integration in GEN AI models

FOUNDATIONAL MODELS:

 Foundational models are models that are already trained in large


amount of data which is serving as a base for various downstream tasks..
It has the ability to handle resources for fine tuning of the model .It has
wide amount of use cases in various fields such as
NLP,Healthcare,Finance ,education ettc…

 FMs ability to handle to the extensive high amount of data and its
capability to fine tune the models made it powerful to enhance its
productivity, automation of complex process and problems to create the
innovative solutions to meet the needs of the users..

AGENTS FOR AMAZON BEDROCK:


1) Agents allow the applications to execute multiple business tasks or
customer needs in Natural language understanding.
2) It doesn’t require high amount of code because it comes with the
understanding of needs ,tools and what and how the execution
should be taken to meet with the needs for the business tasks.
HOW DOES AGENT WORK -WORK FLOW OF THE AGENT:
1)The agent in bedrock decomposes the tasks to get a clear and concise idea to
get what are the job to be done onbehalf of ourselves…
2)Based on the task ,executes action by searching the knowledge bases such as
FMs ,observe and interpret and think what should be done next..(multiple step
process)
3)

You might also like