What you should know - Amazon Bedrock Tutorial

- [Instructor] Here are a few key things to know before we get started. Although I am currently an employee of Amazon, the ideas expressed in this course are my own and not supported by or endorsed by Amazon or AWS. The course assumes you have a basic knowledge of Python. We'll be using it to write some code to interact with the Amazon Bedrock API. You will need an AWS account with access to the AWS console. If you don't have one, set up an AWS account now at aws.amazon.com. Before using Amazon Bedrock, you'll need to request access to the underlying foundation models from the AWS console. We will walk through that process later on in a course. To get resources for this course, visit the GitHub repository for the course using the link shown here. We will be using GitHub Codespaces in this course to make setup of your IDE simpler, and this way, you can use a browser-based IDE that resembles Visual Studio Code for the course with minimal setup needed on your part. To use Codespaces, which is free for this course, you'll need to create a GitHub account. Sign up now for an account at github.com if you do not already have one. This field is constantly evolving. While this is exciting, it can also be challenging. Specifically, the coding libraries we'll be using, like LangChain and underlying APIs, are continuously being updated, which could result in code breakage as these libraries evolve. To minimize this, I have put specific versions of underlying Python libraries and their requirements. You can also try code with the latest versions of these libraries as well. The GitHub repo for the course will be kept up-to-date if issues are identified, so if something does not work as you're coding and following along in a video, please try the latest code from the GitHub repo, and let us know if you encounter any issues with repo code. Foundation models are also constantly being updated with new versions being released monthly. What this means is that the version of a model being used in our code might not match the latest model, or an older version can become deprecated. This is an easy fix. Go to the Amazon Bedrock models page to get the latest model ID and update if needed in the code to prevent any potential issues here. Simply changing one configuration variable will take care of this. We will cover this later in the course. We will be using minimal input and output tokens as part of our interactions with Amazon Bedrock in this course. If your AWS account is still Free tier eligible, you should not have to pay any charges. At most, you must spend about 50 cents US dollars for Amazon Bedrock for this course. Please visit the Amazon Bedrock Pricing page for additional details. We will also be using Amazon Kendra in this course. Once you provision Amazon Kendra by creating an index, you're charge for Amazon Kendra hours from the time an index is created until it's deleted. For the purposes of this course, we'll be using the Developer edition of Amazon Kendra. At the time of this recording, Amazon Kendra provides free usage up to 750 hours for the first 30 days . Since we'll be building a couple connectors and then syncing them on demand, at most, charges for this should be less than $1 US. If you use Amazon Kendra, you'll definitely want to delete your index in the first 30 days of creation to prevent any additional charges. Please visit the Amazon Kendra Pricing page for additional details. We will address all of this in the cleanup video for the course, but to summarize, at most for this course, expect to spend less than $1.50 US on any of the AWS resources used. This can be considerably lower if you're still on the AWS Free tier. With those housekeeping rules in mind, let's set up our AWS environment to get started.