AWS setup - Amazon Bedrock Tutorial

- [Instructor] There are different options to set up security permissions to use Amazon Bedrock. For our course, we're going to keep it simple. We'll create an IAM user that is part of a user group that has access to Amazon Bedrock. Of course, for your specific scenario, you can customize and restrict access depending on your needs. Let's head over to the AWS console. Please note that AWS continues to improve the user experience for its services. For all videos in this course, the AWS console screens may not look exactly the same as when the video is recorded. Please look around if a button or link does not appear in the exact location as described in the video, and you should still be able to follow along. From the top menu, search for and click on IAM. Then click on Users. And Create User. Let's give our user a name. Let's enter Bedrock User. Click on next. Now we're going to associate a user with a group. Click on Create Group. For our group name, let's enter Bedrock. Under permissions policies, start typing Bedrock. We'll then select Amazon Bedrock Full Access, and then click on Create User Group. Next, select the Bedrock User Group and then click on Next. Then finally click on Create User. Now we'll need to get the secret key ID and secret access key for the user. Let's click on Bedrock User. Head over to the security credentials tab, and scroll down and click on Create Access Key. Here you'll want to select other, and then click on Next. Finally, click on Create Access Key. Now, this is the only time that you'll be able to get both the access key and secret access key. Click on the download CSV file and save this locally on your computer. Be sure to keep this file safe. We will refer to this file later when we're setting up our AWS environment. Let's quickly take a look at the structure of this file. This is the file that we downloaded, so I'm going to open this up. And here you'll see it's a comma separated file with the access key ID and secret access key. On the second row, the first value is going to be the access key ID, and after the comma, the second value is going to be your secret access key. Once we've completed setting up our IAM user and group, we're now going to request access to the underlying foundation models we'll be using in Amazon Bedrock. From the AWS console, let's launch Amazon Bedrock. Click on Get Started. Once we're at the Amazon Bedrock page, please ensure that you've selected the AWS region that is closest to you that supports Amazon Bedrock. Only the regions that support Amazon Bedrock will be able to be selected. I'm selecting US West 2. This is important because model access is required to be set up in every region you want to use Amazon Bedrock. From the left hand navigation menu, let's click on model access. Here are the list of models that we can request access to for Amazon Bedrock. You'll notice that the access status is set to be available to request. Now, Claude from Anthropic is a little bit different. We'll be using Claude in this course as it is one of the best FMs for tech summarization and building conversational experiences. We will need to submit a use case before we can request model access. Let's click on manage model access. And first, let's submit the use case for Claude. Click on the submit use case details button. We'll need to enter a few pieces of information. For the purposes of this course, I'm going to be using a fictitious company, so that company is called Kinetico. And here I'm going to copy the URL. Let's head back to the form and you can enter your company name and company website URL as part of your request. But here I'm going to enter my company, the URL, and what industry, you're going to select, of course your appropriate industry. In this case, I'm going to select energy. My intended users would be internal employees and external users. And I am going to select the first use case here for how I will be using my models or first option I should say. Next up, we have to provide a detailed use case and to make things a little bit simpler, I've created some text that you can copy and paste. So if you head over to the course repository in the 00_03 folder, there is claude use case text file. You want to click on it and click on copy. Come back here and we can paste the details in here. And once that's done, go ahead and click on the submit button. So once our use case has been submitted, let's go ahead and request access to the models. So I'm going to select all of the models here from the top left checkbox. Scroll down and we're going to click on request model access. So let's give this a couple seconds here to run. So as you C an see, there are a couple models, specifically Amazon, the Amazon Titan models where access is granted immediately. The other models have in progress as far as the access status, so you'll need to wait maybe a couple minutes for the access to be granted. Generally, access to most models should be granted instantly with the exception of Claude. It may take a few minutes to a couple hours to get access approved. Now that we have our IAM user created with appropriate permission set up and we have requested access to our Amazon bedrock models, next, let's take a look at how we set up our AWS credentials.