<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Amanda Quint on Medium]]></title>
        <description><![CDATA[Stories by Amanda Quint on Medium]]></description>
        <link>https://medium.com/@acquint?source=rss-a1d710e0dbe5------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*agZqeGIA2_qf-Pgcb6ycBw.jpeg</url>
            <title>Stories by Amanda Quint on Medium</title>
            <link>https://medium.com/@acquint?source=rss-a1d710e0dbe5------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 20:56:54 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@acquint/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Exploring Classification Using AWS’s Bedrock and Comprehend]]></title>
            <link>https://medium.com/@acquint/exploring-classification-using-awss-bedrock-and-comprehend-623ca7c2dacc?source=rss-a1d710e0dbe5------2</link>
            <guid isPermaLink="false">https://medium.com/p/623ca7c2dacc</guid>
            <category><![CDATA[women-in-tech]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[web-development]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[aws]]></category>
            <dc:creator><![CDATA[Amanda Quint]]></dc:creator>
            <pubDate>Fri, 01 May 2026 19:30:56 GMT</pubDate>
            <atom:updated>2026-05-01T19:30:56.628Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*9APAgXyiG1G0RWDU" /><figcaption>Photo by <a href="https://unsplash.com/@johnsalzarulo?utm_source=medium&amp;utm_medium=referral">John Salzarulo</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>I was recently handed a classification problem where a user wanted to separate a number of different credit card transactions into internal buckets. There were only a few hundred transactions, and a small number of these had been classified.</p><p>I decided to try an AI-based approach to classify the remaining transactions and to get some more familiarity with using some of AWS’s services.</p><h3>Starting Point</h3><p>I had this data available in CSV format, and for this proof of concept, I used a Python script to manipulate the data into a parseable format and invoke the AWS services using <a href="https://docs.aws.amazon.com/boto3/latest/">boto3</a> .</p><h3>AWS Bedrock</h3><p>My first attempt was to use <a href="https://aws.amazon.com/bedrock/">AWS Bedrock</a> — a service that supplies users with access to many different models from different companies. For this project, I used Bedrock’s invokeMethod to access Anthropic’s small and efficient <a href="https://www.anthropic.com/news/claude-haiku-4-5">Claude Haiku 4.5 </a>model.</p><p>I first wrote a prompt telling the model what information I was giving it and what I would like it to do. I also gave it a transaction to classify, and requested that it return a JSON response containing both its <strong>recommended classification </strong>of the transaction along with its <strong>confidence</strong> in its recommendation.</p><p>This was extremely straightforward using boto3 :</p><pre>bedrock_client = boto3.client(service_name=&quot;bedrock-runtime&quot;, region_name=&quot;us-east-2&quot;)<br>model_id = &quot;us.anthropic.claude-haiku-4-5-20251001-v1:0&quot;<br>body = json.dumps(<br>            {<br>                &quot;anthropic_version&quot;: &quot;bedrock-2023-05-31&quot;,<br>                &quot;max_tokens&quot;: 30,<br>                &quot;messages&quot;: [<br>                    {&quot;role&quot;: &quot;user&quot;, &quot;content&quot;: &lt;PROMPT HERE&gt;}<br>                ],<br>                &quot;temperature&quot;: 0,<br>            }<br>)<br><br>response = bedrock_client.invoke_model(modelId=model_id, body=body)</pre><p>One thing that drove me a little crazy was that no matter how many times I requested a JSON response in the prompt, the response was <strong>always</strong> formatted as Markdown. I finally ended up writing a regex to extract the information that I requested from the Markdown response.</p><p>When I ran my test set through this, the model responded with an overall 95% accuracy, which was great; however, the model was <strong>extremely</strong> confident in itself — rarely returning a confidence score on any particular transaction less than 95%.</p><h3>AWS Comprehend</h3><p>Second, I decided to compare my Bedrock results with results from <a href="https://aws.amazon.com/comprehend/">AWS Comprehend</a>. Comprehend is a machine learning service specifically geared towards natural language processing and classification, so I was eager to see how it would compare.</p><p>The first step here was to create a custom model — I created a special CSV for training. Ideally, I would have had more training data, as I was under the 10-examples-per-classification suggestion, but since this was just an experiment, I went ahead and moved forward with it. Once it finished training, I created a new custom endpoint within Comprehend to access my new model.</p><p>Once this was set up, invoking it in my script was also very simple:</p><pre>comprehend_client = boto3.client(&quot;comprehend&quot;, region_name=&quot;us-east-2&quot;)<br>response = comprehend_client.classify_document(<br>    Text=&lt;transaction to classify&gt;, EndpointArn=&lt;COMPREHEND_ENDPOINT_ARN&gt;<br>)<br></pre><p>This response also contained a classification recommendation and a confidence score.</p><p>One thing that I appreciated about using Comprehend is that the confidence score felt more real. Whereas Claude told me that its confidence was basically always 95%, Comprehend returned values more like 76.27%, with a unique value for each transaction. Whether its confidence score really was better or not, it made me feel less like it was lying to me.</p><h3>Conclusions</h3><p>I was actually shocked that both methods returned the same percentage of correct results. When comparing, they didn’t always get the same transactions misclassified either — each one did better on a different set of transactions.</p><p>Obviously, to run this experiment at a production scale, I would need both <strong>more data </strong>and<strong> more experimentation</strong>: different and/or custom models, different prompts, feeding in additional information (i.e., transaction amount), but I felt like this was an interesting place to start. When dealing with services like this, there can be infinite experimentation: tweaking and retraining prompts and models until you find an optimized experience.</p><p>As an additional note, just playing around with these services was surprisingly expensive: I racked up over $6 on Bedrock and over $12 on Comprehend, so that is definitely something to take into consideration when exploring these services (or running them at a production scale!).</p><p><em>I enjoy writing about software development, project management, and my journey in the AWS Cloud. If you’d like to read more, please consider following me here on Medium or </em><a href="https://www.linkedin.com/in/amanda-quint-6b699878/"><em>LinkedIn</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=623ca7c2dacc" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Reading 80 Books in 2025 — with some commentary]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/illumination/reading-80-books-in-2025-with-some-commentary-225ca694a93a?source=rss-a1d710e0dbe5------2"><img src="https://cdn-images-1.medium.com/max/2600/0*BLtKSscT5pKXB43V" width="5294"></a></p><p class="medium-feed-snippet">My Annual Reading Challenge Summed Up</p><p class="medium-feed-link"><a href="https://medium.com/illumination/reading-80-books-in-2025-with-some-commentary-225ca694a93a?source=rss-a1d710e0dbe5------2">Continue reading on ILLUMINATION »</a></p></div>]]></description>
            <link>https://medium.com/illumination/reading-80-books-in-2025-with-some-commentary-225ca694a93a?source=rss-a1d710e0dbe5------2</link>
            <guid isPermaLink="false">https://medium.com/p/225ca694a93a</guid>
            <category><![CDATA[reading]]></category>
            <category><![CDATA[goals]]></category>
            <category><![CDATA[self-improvement]]></category>
            <category><![CDATA[books]]></category>
            <category><![CDATA[new-years-resolutions]]></category>
            <dc:creator><![CDATA[Amanda Quint]]></dc:creator>
            <pubDate>Fri, 23 Jan 2026 00:06:18 GMT</pubDate>
            <atom:updated>2026-01-23T00:06:18.208Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Debugging S3’s Pre-signed URLs with Boto3 and Python]]></title>
            <link>https://medium.com/@acquint/debugging-s3s-pre-signed-urls-with-boto3-and-python-26af316d7f9a?source=rss-a1d710e0dbe5------2</link>
            <guid isPermaLink="false">https://medium.com/p/26af316d7f9a</guid>
            <category><![CDATA[women-in-tech]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[software]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[Amanda Quint]]></dc:creator>
            <pubDate>Thu, 27 Mar 2025 21:56:48 GMT</pubDate>
            <atom:updated>2025-03-27T21:56:48.808Z</atom:updated>
            <content:encoded><![CDATA[<h4>Dealing with Unexpected Region Behavior</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*RBvAysDMTKJJW0FT" /><figcaption>Photo by <a href="https://unsplash.com/@pedrotheartist?utm_source=medium&amp;utm_medium=referral">Pedro Forester Da Silva</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p><a href="https://docs.aws.amazon.com/s3/">Amazon’s Simple Storage Service (S3)</a> is an extremely robust cloud storage solution, and it has a ton of settings and functionality: including the ability to create a “pre-signed” URL that gives others permission to access objects in your bucket for a given time.</p><p>This story documents that functionality and an unexpected quirk I ran into while implementing it.</p><h3><strong>Background</strong></h3><p>I have a small application that allows a user to upload a file to an S3 bucket that the application owns. This action kicks off a series of processing events on the file.</p><p>When a user first initiates this process, a Lambda function hosting a Python API is hit and creates a <a href="https://docs.aws.amazon.com/AmazonS3/latest/userguide/using-presigned-url.html">“Pre-signed” POST URL</a>, which gives the application a few seconds to upload the file.</p><p>I originally wrote the code for this in my development environment, which is located in us-east-1 — however, when I deployed the code to a new environment located in us-east-2 the application wouldn’t work, although I was deploying the same code in both places.</p><p>After a lot of trial and error (and googling), I finally figured out what was happening.</p><h3>Reproducing the Issue</h3><p>To reproduce this, I created the following resources:</p><ol><li>testpresignedpostbucket — a private S3 bucket in us-east-1</li><li><strong>testS3presignedpostUSEAST1 </strong>— a Lambda function deployed in us-east-1 with a role that gives it permission to read and write to S3</li><li>testpresignedpostbucketeast2 — a private S3 bucket in us-east-2</li><li><strong>testS3presignedpostUSEAST2 </strong>— a Lambda function deployed in us-east-2 with a role that gives it permission to read and write to S3</li></ol><p>Both Lambdas are running Python 3.12, are set to be invoked using a <a href="https://docs.aws.amazon.com/lambda/latest/dg/urls-invocation.html">Function URL</a>, and have the same small snippet of code deployed; the only change being the corresponding bucket name:</p><pre>import json<br>import boto3<br><br>bucket = &#39;testpresignedpostbucket&#39; # append &quot;east2&quot; in that region<br><br>def lambda_handler(event, context):<br>    s3_client = boto3.client(&#39;s3&#39;)<br>    return s3_client.generate_presigned_post(bucket, &#39;test.txt&#39;, ExpiresIn=60)</pre><p>This code uses boto3 to generate a pre-signed post, which will pass back a URL and the form fields needed to allow a user to upload an object to the bucket key “test.txt” for the next 60 seconds. I’m using the generate_presigned_post method because users are only writing to the bucket, they never read from it.</p><p>Finally, I also created a small text file to try to upload: test.txt , which only contained the words “Hello World!”</p><h3>Testing the Resources</h3><blockquote>Note: The resources have all been deleted as of my writing this, so the following curl calls are for examples only.</blockquote><p>I hit my Lambda in us-east-1 with the following curl command:</p><pre>curl --request GET \<br>  --url https://isyo6u4s6j2mb2lfngzmbdas4u0pzkll.lambda-url.us-east-1.on.aws/</pre><p>In response, I got the following:</p><pre>{<br> &quot;fields&quot;: {<br>  &quot;signature&quot;: &quot;********&quot;,<br>  &quot;AWSAccessKeyId&quot;: &quot;********&quot;,<br>  &quot;x-amz-security-token&quot;: &quot;********&quot;,<br>  &quot;key&quot;: &quot;test.txt&quot;,<br>  &quot;policy&quot;: &quot;********==&quot;<br> },<br> &quot;url&quot;: &quot;https:\/\/testpresignedpostbucket.s3.amazonaws.com\/&quot;<br>}</pre><p>Based on this response, I wrote the code in my application to use this information to upload my test.txt file:</p><pre>import requests<br><br>def get_presigned_url(https://medium.com/function_url):<br>    response = requests.get(function_url)<br>    response.raise_for_status()<br>    upload_file(response.json())<br><br>def upload_file(presigned_url_fields):<br>    payload = {<br>        &quot;signature&quot;: presigned_url_fields[&quot;fields&quot;].get(&quot;signature&quot;),<br>        &quot;AWSAccessKeyId&quot;: presigned_url_fields[&quot;fields&quot;].get(&quot;AWSAccessKeyId&quot;),<br>        &quot;policy&quot;: presigned_url_fields[&quot;fields&quot;].get(&quot;policy&quot;),<br>        &quot;x-amz-security-token&quot;: presigned_url_fields[&quot;fields&quot;].get(<br>            &quot;x-amz-security-token&quot;<br>        ),<br>        &quot;key&quot;: &quot;test.txt&quot;,<br>    }<br>    with open(&quot;test.txt&quot;, &quot;rb&quot;) as f:<br>        response = requests.post(<br>            presigned_url_fields[&quot;url&quot;], data=payload, files={&quot;file&quot;: f}<br>        )<br>        response.raise_for_status()<br>        print(response.status_code)</pre><p>This worked as expected, I got an HTTP 204 in response, and my file ended up where I expected in my bucket:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*s7Lp_AIDJE-FiI1vEkCX_A.png" /><figcaption>Test.txt was successfully uploaded to S3</figcaption></figure><h4>In us-east-2</h4><p>However, as soon as I deployed this code to us-east-2 my Python script started throwing an HTTP 403 Client Error: Forbidden for url . My initial thought was that my bucket permissions were somehow misconfigured, but that didn’t make sense, as I had deployed the same CDK stack in both regions.</p><p>Curling the us-east-2 Lambda revealed a <em>different set of keys, even though it was running the same code:</em></p><pre>curl --request GET \<br>  --url https://yyji6kiile2t74orqi6bxppjxq0ybhei.lambda-url.us-east-2.on.aws/</pre><pre>{<br> &quot;fields&quot;: {<br>  &quot;x-amz-date&quot;: &quot;********&quot;,<br>  &quot;x-amz-signature&quot;: &quot;********&quot;,<br>  &quot;x-amz-security-token&quot;: &quot;********&quot;,<br>  &quot;key&quot;: &quot;test.txt&quot;,<br>  &quot;x-amz-algorithm&quot;: &quot;AWS4-HMAC-SHA256&quot;,<br>  &quot;x-amz-credential&quot;: &quot;********&quot;,<br>  &quot;policy&quot;: &quot;********==&quot;<br> },<br> &quot;url&quot;: &quot;https:\/\/testpresignedpostbucketeast2.s3.amazonaws.com\/&quot;<br>}</pre><p>Since this was <strong>the same code </strong>deployed (via Amazon CDK) in two different regions, I was really confused!</p><h3>Solution</h3><p>It turns out that this issue is due to the AWS Signature Version.</p><p>AWS S3 uses Signature Version 4 in all regions. Regions created <em>after</em> January 30th, 2014 also default to using Version 4. However, regions created <em>before </em>January 30th, 2014 still support Version 2. See <a href="https://docs.aws.amazon.com/AmazonS3/latest/API/sig-v4-authenticating-requests.html">AWS’s documentation on Authenticating Requests</a> for more information.</p><p>us-east-1 was AWS’s first region and launched in 2006. <br>us-east-2 didn’t launch until <em>ten years later </em>in 2016! (After the signature version change date)</p><h4>Code Change</h4><p>I found a <a href="https://github.com/boto/boto3/issues/2606#issuecomment-701587119">GitHub issue</a> that explained my problem: <strong>Botocore also defaults to using Version 2 while generating pre-signed posts in </strong><strong>us-east-1 </strong>— so to make this work a simple code change to specify the signature version was needed:</p><pre>import json<br>import boto3<br>from botocore.client import Config<br><br>bucket = &#39;testpresignedpostbucket&#39;<br><br>def lambda_handler(event, context):<br>    s3_client = boto3.client(&#39;s3&#39;, config=Config(signature_version=&quot;s3v4&quot;))<br>    return s3_client.generate_presigned_post(bucket, &#39;test.txt&#39;, ExpiresIn=60)</pre><p>This results in the following response, with all the expected fields that we were seeing in us-east-2 :</p><pre>{<br> &quot;fields&quot;: {<br>  &quot;x-amz-date&quot;: &quot;********&quot;,<br>  &quot;x-amz-signature&quot;: &quot;********&quot;,<br>  &quot;x-amz-security-token&quot;: &quot;********&quot;,<br>  &quot;key&quot;: &quot;test.txt&quot;,<br>  &quot;x-amz-algorithm&quot;: &quot;AWS4-HMAC-SHA256&quot;,<br>  &quot;x-amz-credential&quot;: &quot;********&quot;,<br>  &quot;policy&quot;: &quot;********==&quot;<br> },<br> &quot;url&quot;: &quot;https:\/\/testpresignedpostbucket.s3.amazonaws.com\/&quot;<br>}</pre><p>I would also like to note that if I had not been explicit in my Python script, I wouldn’t have run into this issue. However, the boto3 documentation for generate_presigned_post is a little bit vague on the fields dictionary — I could have used a bit more documentation.</p><pre>def get_presigned_https://medium.com/url(url):<br>    response = requests.get(url)<br>    response.raise_for_status()<br>    upload_file(response.json())<br><br><br>def upload_file(presigned_url_fields):<br>    with open(&quot;test.txt&quot;, &quot;rb&quot;) as f:<br>        response = requests.post(<br>            presigned_url_fields[&quot;url&quot;],<br>            data=presigned_url_fields[&quot;fields&quot;],<br>            files={&quot;file&quot;: f},<br>        )<br>        response.raise_for_status()<br>        print(response.status_code)</pre><p>Ultimately, I feel like it’s better to use the newer signature version in both cases.</p><h3>Another Gotcha — Order of Fields Keys Matter!</h3><p>While writing this story, I stumbled upon another thing I didn’t expect: the order of the fields seems to matter.</p><p>Using Insomnia, I was testing uploading the file to the pre-signed URL and encountered the following 400 Bad Request error:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tvOwtB63xVR8UQKyr7zMUw.png" /><figcaption>I received a 400 Bad Request when the fields were out of order</figcaption></figure><p>This was fixed by making sure that the key field came before the file :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rHqKo3quI5a6nm5gU7iMyw.png" /><figcaption>The name information sent in a different order results in a sucessful 204 response</figcaption></figure><p>This is a minor issue, but it gave me a moment of confusion!</p><h3>Conclusion</h3><p>When I deploy code in different regions, I typically expect the code to act the same way. I know that brand-new features may only be available in certain regions, but for heavily used services like Lambda and S3, I wasn’t expecting to find any quirks. It took me a couple of hours to track down what was happening in this case, so I hope this article can perhaps save someone else some time in the future!</p><p><em>I enjoy writing about software development, project management, and my journey in the AWS Cloud. If you’d like to read more, please consider following me here on Medium or </em><a href="https://www.linkedin.com/in/amanda-quint-6b699878/"><em>LinkedIn</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=26af316d7f9a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Reading 80 Books in 2024 — with some commentary]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/illumination/reading-80-books-in-2024-with-some-commentary-f2db9860a0df?source=rss-a1d710e0dbe5------2"><img src="https://cdn-images-1.medium.com/max/2600/0*euzu1xHQ-lpeJmDc" width="6000"></a></p><p class="medium-feed-snippet">My Annual Reading Challenge Summed Up</p><p class="medium-feed-link"><a href="https://medium.com/illumination/reading-80-books-in-2024-with-some-commentary-f2db9860a0df?source=rss-a1d710e0dbe5------2">Continue reading on ILLUMINATION »</a></p></div>]]></description>
            <link>https://medium.com/illumination/reading-80-books-in-2024-with-some-commentary-f2db9860a0df?source=rss-a1d710e0dbe5------2</link>
            <guid isPermaLink="false">https://medium.com/p/f2db9860a0df</guid>
            <category><![CDATA[reading]]></category>
            <category><![CDATA[books]]></category>
            <category><![CDATA[goals]]></category>
            <category><![CDATA[self-improvement]]></category>
            <category><![CDATA[book-review]]></category>
            <dc:creator><![CDATA[Amanda Quint]]></dc:creator>
            <pubDate>Fri, 17 Jan 2025 05:10:25 GMT</pubDate>
            <atom:updated>2025-01-17T05:10:25.087Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Automatically Delete Objects in your S3 Bucket using AWS CDK]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://aws.plainenglish.io/automatically-delete-objects-in-your-s3-bucket-using-aws-cdk-640941a368d3?source=rss-a1d710e0dbe5------2"><img src="https://cdn-images-1.medium.com/max/2600/0*Tt0fmce06rappqLd" width="3449"></a></p><p class="medium-feed-snippet">Set up object lifecycles using Infrastructure as Code (IAC)</p><p class="medium-feed-link"><a href="https://aws.plainenglish.io/automatically-delete-objects-in-your-s3-bucket-using-aws-cdk-640941a368d3?source=rss-a1d710e0dbe5------2">Continue reading on AWS in Plain English »</a></p></div>]]></description>
            <link>https://aws.plainenglish.io/automatically-delete-objects-in-your-s3-bucket-using-aws-cdk-640941a368d3?source=rss-a1d710e0dbe5------2</link>
            <guid isPermaLink="false">https://medium.com/p/640941a368d3</guid>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[web-development]]></category>
            <category><![CDATA[women-in-tech]]></category>
            <category><![CDATA[programming]]></category>
            <dc:creator><![CDATA[Amanda Quint]]></dc:creator>
            <pubDate>Mon, 25 Mar 2024 08:07:25 GMT</pubDate>
            <atom:updated>2024-03-25T08:07:25.675Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Creating an Amazon EventBridge Scheduler with CDK L2 Constructs (Updated for 2024)]]></title>
            <link>https://aws.plainenglish.io/creating-an-amazon-eventbridge-scheduler-with-cdk-l2-constructs-updated-for-2024-6d64166869c3?source=rss-a1d710e0dbe5------2</link>
            <guid isPermaLink="false">https://medium.com/p/6d64166869c3</guid>
            <category><![CDATA[web-development]]></category>
            <category><![CDATA[women-in-tech]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[Amanda Quint]]></dc:creator>
            <pubDate>Thu, 21 Mar 2024 13:25:04 GMT</pubDate>
            <atom:updated>2024-03-21T13:25:04.107Z</atom:updated>
            <content:encoded><![CDATA[<h4>Schedule your workflows using Infrastructure as Code (IAC).</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*KLoYbsKoc-fekEh5" /><figcaption>Photo by <a href="https://unsplash.com/@sonjalangford?utm_source=medium&amp;utm_medium=referral">Sonja Langford</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Amazon’s E<a href="https://aws.amazon.com/eventbridge/scheduler/">ventBridge Scheduler</a> was released in 2022, making it easy to schedule workflows on a one-time or recurring basis. In March 2023, <a href="https://aws.plainenglish.io/creating-an-amazon-eventbridge-scheduler-with-cdk-5fca2086b154">I wrote an article</a> explaining the benefits of using EventBridge Scheduler, and how to implement it using the constructs the <a href="https://aws.amazon.com/cdk/">AWS Cloud Development Kit (CDK)</a> had available at the time.</p><p>While the <a href="https://aws.plainenglish.io/creating-an-amazon-eventbridge-scheduler-with-cdk-5fca2086b154">original article</a> still has up-to-date information about why you should use EventBridge Scheduler and the base-level (L1) CfnResource implementation will still work, this article covers how to implement EventBridge Scheduler in CDK using the newer custom (L2) constructs.</p><h3>Step 1: Install Packages</h3><p>At the time of writing (March 2024), the CDK packages needed for EventBridge Scheduler are still in alpha. This means that they could still be missing some functionality or experience non-backward compatible changes. Keep this in mind before you use them!</p><blockquote>Note: Depending on the version of CDK you’re using, you may need to update before installing these packages. Check your version if you get an error on installation!</blockquote><p>The first package is for the scheduler itself. The official documentation for this package can be found <a href="https://docs.aws.amazon.com/cdk/api/v2/docs/aws-scheduler-alpha-readme.html">here</a>. To install it:</p><pre>npm install &quot;@aws-cdk/aws-scheduler-alpha&quot;<br></pre><p>The second package is for the Scheduler’s targets. The official documentation for this package can be found <a href="https://docs.aws.amazon.com/cdk/api/v2/docs/aws-scheduler-targets-alpha-readme.html#invoke-a-lambda-function">here</a>. To install it:</p><pre>npm install &quot;@aws-cdk/aws-scheduler-targets-alpha&quot;</pre><p>Note that as of the time of writing (March 2024), the following targets are supported in this construct:</p><ul><li>Lambda</li><li>Step Functions</li><li>CodeBuild</li><li>SQS</li><li>SNS</li><li>EventBridge (put events)</li><li>Amazon Inspector</li><li>Kinesis Data Streams</li><li>Data Firehose (still using Kinesis in the target name)</li><li>CodePipeline</li><li>SageMaker</li></ul><h3>Step 2: Update Code</h3><p>Prior to the L2 constructs being available, the way to implement an EventBridge scheduler was by using CfnResource — <a href="https://docs.aws.amazon.com/cdk/v2/guide/constructs.html">an L1 construct</a>. L1 constructs are generated directly from AWS CloudFormation and are available soon after new functionality is released, whereas L2 constructs are built by AWS’s CDK team.</p><p>To continue the example used in my previous article, the CfnResource we defined looked like this:</p><pre>// EventBridge Scheduler using CfnResource<br>const myScheduler = new cdk.CfnResource(this, &quot;myScheduler&quot;, {<br> type: &quot;AWS::Scheduler::Schedule&quot;,<br> properties: {<br>  Name: &quot;myTestEventBridgeScheduler&quot;,<br>  Description: &quot;Runs a lambda every weekday at 6 AM EST&quot;,<br>  FlexibleTimeWindow: { Mode: &quot;OFF&quot; },<br>  ScheduleExpression: &quot;cron(0 6 ? * MON-FRI *)&quot;,<br>  ScheduleExpressionTimezone: &quot;America/New_York&quot;,<br>  Target: {<br>    Arn: myTestLambda.functionArn,<br>    RoleArn: schedulerRole.roleArn,<br>  },<br>},</pre><p>This Scheduler invoked a Lambda function to run every weekday at 6 am EST. Rewriting this same Scheduler to use the new L2 constructs looks like this:</p><pre>import * as scheduler from &quot;@aws-cdk/aws-scheduler-alpha&quot;;<br>import * as schedulerTargets from &quot;@aws-cdk/aws-scheduler-targets-alpha&quot;;<br><br>// EventBridge Scheduler using L2 Constructs<br>const myScheduler = new scheduler.Schedule(this, &quot;myScheduler&quot;, {<br>  scheduleName: &quot;myTestEventBridgeScheduler&quot;,<br>  description: &quot;Runs a lambda every weekday at 6am EST&quot;,<br>  schedule: scheduler.ScheduleExpression.cron({<br>      minute: &quot;0&quot;,<br>      hour: &quot;6&quot;,<br>      month: &quot;*&quot;,<br>      year: &quot;*&quot;,<br>      weekDay: &quot;MON-FRI&quot;,<br>      timeZone: cdk.TimeZone.AMERICA_NEW_YORK,<br>  }),<br>  target: new schedulerTargets.LambdaInvoke(myTestLambda, <br>                                            {role: schedulerRole}),<br>},</pre><p>Note that we were able to drop any reference to the FlexibleTimeWindow , as it defaults to off in the L2 construct. Not passing a role in our LambdaInvoke would create a role for us, and there’s some enhanced functionality we’re not using here, such as passing a custom payload and a dead-letter queue. Check out the LambdaInvoke <a href="https://docs.aws.amazon.com/cdk/api/v2/docs/aws-scheduler-targets-alpha-readme.html#invoke-a-lambda-function">documentation</a> for more information.</p><p>While both of these examples end up producing the same thing, I typically find that reading L2 constructs is easier as they’re more straightforward about what they’re doing. (There is also some nice auto-complete and documentation functionality while writing them in TypeScript using VSCode!)</p><h3>Other CDK Resources</h3><p>Here are a few other resources that may come in handy when working with AWS CDK:</p><ul><li><a href="https://docs.aws.amazon.com/cdk/api/v2/docs/aws-construct-library.html">The official AWS Reference</a> contains documentation on all the constructs</li><li><a href="https://constructs.dev/">Construct Hub </a>contains other open-source constructs</li><li>I also like to write about problems I’m able to solve with CDK, so please <a href="https://medium.com/@acquint/list/aws-dbbe2c090c57">check out my list of AWS-related articles</a></li></ul><p><em>I enjoy writing about software development, project management, and my journey in the AWS Cloud. If you’d like to read more, please consider following me here on Medium, </em><a href="https://twitter.com/acquint"><em>Twitter</em></a><em>, or </em><a href="https://www.linkedin.com/in/amanda-quint-6b699878/"><em>LinkedIn</em></a><em>.</em></p><h3>In Plain English 🚀</h3><p><em>Thank you for being a part of the </em><a href="https://plainenglish.io/"><strong><em>In Plain English</em></strong></a><em> community! Before you go:</em></p><ul><li>Be sure to <strong>clap</strong> and <strong>follow</strong> the writer ️👏<strong>️️</strong></li><li>Follow us: <a href="https://twitter.com/inPlainEngHQ"><strong>X</strong></a><strong> | </strong><a href="https://www.linkedin.com/company/inplainenglish/"><strong>LinkedIn</strong></a><strong> | </strong><a href="https://www.youtube.com/channel/UCtipWUghju290NWcn8jhyAw"><strong>YouTube</strong></a><strong> | </strong><a href="https://discord.gg/in-plain-english-709094664682340443"><strong>Discord</strong></a><strong> | </strong><a href="https://newsletter.plainenglish.io/"><strong>Newsletter</strong></a></li><li>Visit our other platforms: <a href="https://stackademic.com/"><strong>Stackademic</strong></a><strong> | </strong><a href="https://cofeed.app/"><strong>CoFeed</strong></a><strong> | </strong><a href="https://venturemagazine.net/"><strong>Venture</strong></a><strong> | </strong><a href="https://blog.cubed.run/"><strong>Cubed</strong></a></li><li>More content at <a href="https://plainenglish.io/"><strong>PlainEnglish.io</strong></a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6d64166869c3" width="1" height="1" alt=""><hr><p><a href="https://aws.plainenglish.io/creating-an-amazon-eventbridge-scheduler-with-cdk-l2-constructs-updated-for-2024-6d64166869c3">Creating an Amazon EventBridge Scheduler with CDK L2 Constructs (Updated for 2024)</a> was originally published in <a href="https://aws.plainenglish.io/">AWS in Plain English</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My First Experience Building a Customized Chatbot in AWS]]></title>
            <link>https://levelup.gitconnected.com/my-first-experience-building-a-customized-chatbot-in-aws-3bb07f4571e7?source=rss-a1d710e0dbe5------2</link>
            <guid isPermaLink="false">https://medium.com/p/3bb07f4571e7</guid>
            <category><![CDATA[women-in-tech]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Amanda Quint]]></dc:creator>
            <pubDate>Thu, 25 Jan 2024 21:53:41 GMT</pubDate>
            <atom:updated>2024-01-25T21:53:41.156Z</atom:updated>
            <content:encoded><![CDATA[<h4>Using Amazon Bedrock, Titan Models, and SageMaker Studio</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*viMUN9Cd428S6z2t" /><figcaption>Photo by <a href="https://unsplash.com/@ekrull?utm_source=medium&amp;utm_medium=referral">Eric Krull</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>Introduction</h3><p>In November 2023, AWS announced a lot of new features and capabilities for Generative AI. These included the <a href="https://aws.amazon.com/bedrock/titan/">Amazon Titan models</a> available through <a href="https://aws.amazon.com/bedrock/">Amazon Bedrock</a>.</p><p>I am not a data scientist or machine learning expert, although I have enjoyed <a href="https://medium.com/gitconnected/passing-the-aws-machine-learning-speciality-exam-8c90d87bb011">learning more about machine learning</a> recently. This story is not a technical deep dive or tutorial, but rather the experience I went through using these new services for the first time, the hiccups and issues I ran into along the way, and what I ended up with at the end.</p><p>If you are looking for a tutorial, I highly recommend this <a href="https://catalog.us-east-1.prod.workshops.aws/workshops/a4bdb007-5600-4368-81c5-ff5b4154f518/en-US">Amazon Bedrock Workshop</a> and its corresponding <a href="https://github.com/aws-samples/amazon-bedrock-workshop/tree/main?tab=readme-ov-file">Github repository</a> — this is what I used to build my chatbot.</p><h4>My Data Set and Idea</h4><p>I have been doing consulting and contracting work for <a href="https://www.greatgame.com/community">The Great Game of Business Community</a>, a professional community built around connecting and educating practitioners on the business methodologies known as the “Great Game of Business™”. They were kind enough to share PDF versions of 3 of their books, as well as over 100 case studies, “MiniGame™” workshops, training tools, and other supplementary resources with me to produce a knowledge base to train my model on.</p><p>Therefore, my chatbot is trained to answer questions about the Great Game™, and how to implement it.</p><h3>First Steps</h3><p>The first thing that I had to do was make sure that my AWS account had the proper permissions to use the models, both in IAM and in my account as a whole.</p><p>After reading through the different models available in <a href="https://aws.amazon.com/bedrock/">Bedrock</a>, I decided to go with the <a href="https://aws.amazon.com/bedrock/titan/">Titan</a> family of models, which were created by AWS. Requesting access to Titan was easy, and access was granted almost immediately.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1uXWoJztr7ElNxYbTgjn6w.png" /><figcaption>Access to Amazon’s Titan Models</figcaption></figure><p>The second thing that I needed to do was get <a href="https://aws.amazon.com/pm/sagemaker/?gclid=Cj0KCQiA-62tBhDSARIsAO7twbbJO9-bQd67FyhcRznhK6tjkoPdP2mXPD8zuBP9qbuHZN4T4NduFI4aAl7REALw_wcB&amp;trk=b6c2fafb-22b1-4a97-a2f7-7e4ab2c7aa28&amp;sc_channel=ps&amp;ef_id=Cj0KCQiA-62tBhDSARIsAO7twbbJO9-bQd67FyhcRznhK6tjkoPdP2mXPD8zuBP9qbuHZN4T4NduFI4aAl7REALw_wcB:G:s&amp;s_kwcid=AL!4422!3!651751060695!e!!g!!sagemaker%20studio!19852662230!145019226137">SageMaker Studio</a> up and running. I believe that due to the age of my AWS account, I did not have any SageMaker Endpoints available in my service limits. I had to request a limit increase as a support case. AWS was happy to bump my limit up to 5, but it took about 48 hours, which felt like a long time to wait while I was excited about building something!</p><p>Since I was following the <a href="https://catalog.us-east-1.prod.workshops.aws/workshops/a4bdb007-5600-4368-81c5-ff5b4154f518/en-US/20-intro/21-environmentsetup">workshop</a> mentioned above, I also used SageMaker Studio Classic. This workshop also provided a notebook that contained requirements to run the following lab notebooks, and set up boto3 , awscli , langchain , etc.</p><p>Finally, I ensured that I had a budget set on my AWS account. Any time you start playing with AI or ML-related services, costs can grow quickly!</p><h3>My Notebook</h3><p>None of the labs in the workshop followed exactly what I wanted to do, so I pieced them together into my own Frankenstein-style notebook. While this is probably not as educational as researching and writing the code from scratch, this allowed me to get something working quickly, and over the hump of the learning curve that comes along with working with unfamiliar libraries and services. I modified the code to produce the following:</p><blockquote>Note: As this was not originally my code, please check out the <a href="https://github.com/aws-samples/amazon-bedrock-workshop/tree/main?tab=readme-ov-file">Github repository here</a>. I pulled all my resources from this repository.</blockquote><h4>Step 1: Set up a New boto3 Bedrock client</h4><p>The first thing we had to do was make sure that we were able to communicate with Bedrock:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cq8P0VLZOQEJcbA8DnHAEw.png" /><figcaption>Creating a new boto3 client for Bedrock</figcaption></figure><h4>Step 2: Prepping My Data</h4><p>Since all my PDFs didn’t really add up to much space, I loaded them into a local directory in the SageMaker notebook and read from that directory (larger datasets might want to utilize s3). I used langchain to split the documents into chunk sizes of 2000, based on the lab’s suggestions, but if this was something that I was going to build as a serious project, I would want to optimize with these numbers.</p><h4>Step 3: Creating OpenSearch Vector Database</h4><p><a href="https://aws.amazon.com/opensearch-service/">OpenSearch</a> is AWS’s version of Elasticsearch, and <a href="https://aws.amazon.com/blogs/big-data/amazon-opensearch-services-vector-database-capabilities-explained/">recently it gained the ability to operate as a vector database</a>. In the code, this is as simple as passing type=&#39;VECTORSEARCH&#39; when you create the collection, but as I learned later, the <a href="https://aws.amazon.com/opensearch-service/pricing/?refid=b6c2fafb-22b1-4a97-a2f7-7e4ab2c7aa28">price</a> of OpenSearch Serverless makes it prohibitive for most personal projects such as this one. (More on this below)</p><p>It took about 5 minutes to build a new collection.</p><h4>Step 4: Instantiating the LLM and Embeddings</h4><p>I created the LLM using Titan and the Bedrock Embeddings that we would use to populate the vector search:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/922/1*4DBRKlD_X14NM_dxKpxXTw.png" /><figcaption>Creating the LLM and the BedrockEmbeddings</figcaption></figure><h4>Step 5: Loading the Documents into OpenSearch</h4><p>This took the docs that I had prepped in Step 2, and loaded them into the OpenSearch Vector Search, using the Bedrock Embeddings that I had just created.</p><p>This also took about 5 minutes.</p><h4>Step 6: Chatbot UX</h4><p>The workshop included a class for a small UX for conversing with the chatbot. I set up this UX pointed to the LLM that I had created and the vector search as my retriever.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/806/1*9pQb9XkwIjmyVBvuPH3n8A.png" /><figcaption>The chatbot is waiting to chat!</figcaption></figure><p>And that was it! The chatbot was up and ready to chat!</p><h3>Chatting</h3><p>Here are some of the things that I asked my “Great Game Chatbot” about, and the responses it gave me. Overall, I was impressed, but obviously, I had to keep the conversation to the topic at hand.</p><h4>What is the Great Game of Business all about?</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cPyKu-f4Uf7x9Jy6lR6Faw.png" /><figcaption>AI: The Great Game of Busines is an open-book management system that takes the elements needed to win any game and applies them to the art and science of growing a successful business.</figcaption></figure><h4>What are the top 3 most important things in the Great Game?</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*W87YnSxN_4lEnw1oM_jcqw.png" /><figcaption>AI: The top three most important things in the great game of business are: 1. When you appeal to the highest level of thinking, you get the highest level of performance. 2. Business is a game of fractions. The more employees know about your company, the more capable they are of making it better. 3. Numbers are not a substitute for leadership. Numbers tell you where the problems are- and how worried you should be.</figcaption></figure><h4>What should I do to start implementing the Great Game at my business?</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hxSBUBMpkUY-m0FI495MtQ.png" /><figcaption>AI: To start implementing the Great Game at your business, you and your managers must be willing to look at business through a whole new prism and sharpen two critical skills: teaching and motivating employees. Open the books and share the big picture, show them how the business works and what is critical to success, and engage them by providing them with a Stake in the Outcome. You can also use a practitioner coach to provide guidance and support during the implementation process.</figcaption></figure><h4>How do I bake a pie?</h4><p>Sadly, it did not know everything.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/944/1*rDwFJ-UEmXXgnNzypQXpEA.png" /><figcaption>AI: Sorry, I don’t know, as the answer was not found in the context.</figcaption></figure><h3>Final Steps</h3><p>After having fun asking the chatbot everything I could think of, the final steps were to delete the OpenSearch collection and to shut down the SageMaker Studio instance.</p><h4>Costs</h4><p>Turns out that my chatbot was expensive, even to play with: I only left it running for a couple of days, but I managed to rack up over $20 in charges! (I<em> highly suggest setting up a Budget in your AWS account before you start playing with these services</em>.) The highest charge was for my use of OpenSearch, which was over $17.</p><p>I was also surprised at the cost of SageMaker Studio, which was almost $6 on its own. However — Bedrock was actually much cheaper than I was expecting: my charges only came out to $0.35.</p><h3>Final Thoughts</h3><ol><li>This was fun, but I’m not sure of the practicality for most use cases. However, GenAI-related tech is currently exploding in popularity, so there are use cases out there!</li><li>If there was a business need for this, I would have to figure out a cheaper Vector database solution than OpenSearch. I know they’re out there, but the way that OpenSearch plugged in made it extremely easy.</li><li>Even though The Great Game was kind enough to share what felt like a LOT of resources with me, I actually think I need even more to make a more robust solution.</li><li>If I did this as a serious project, I would need to do a lot more research. Was Titan the best choice for this? Should I tweak my learning parameters? Did I chunk my documents up optimally? I feel like using ML involves a lot of trial and error, and this was no different.</li></ol><h3>Additional Resources</h3><ul><li>As noted a couple of times, I highly recommend checking out this <a href="https://catalog.us-east-1.prod.workshops.aws/workshops/a4bdb007-5600-4368-81c5-ff5b4154f518/en-US">workshop</a> for more Bedrock tutorials!</li><li>If you’re interested in learning more about The Great Game of Business check them out <a href="https://www.greatgame.com/">here</a>!</li></ul><p><em>I enjoy writing about software development, project management, and my journey in the AWS Cloud. If you’d like to read more, please consider following me here on Medium, </em><a href="https://twitter.com/acquint"><em>Twitter</em></a><em>, or </em><a href="https://www.linkedin.com/in/amanda-quint-6b699878/"><em>LinkedIn</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3bb07f4571e7" width="1" height="1" alt=""><hr><p><a href="https://levelup.gitconnected.com/my-first-experience-building-a-customized-chatbot-in-aws-3bb07f4571e7">My First Experience Building a Customized Chatbot in AWS</a> was originally published in <a href="https://levelup.gitconnected.com/">Level Up Coding</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Compressing Large Payloads for HTTP APIs in API Gateway]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://aws.plainenglish.io/compressing-large-payloads-for-http-apis-in-api-gateway-831d8f63d751?source=rss-a1d710e0dbe5------2"><img src="https://cdn-images-1.medium.com/max/2600/0*ZolsH0Gt1gAkBBpX" width="3031"></a></p><p class="medium-feed-snippet">Using Python and Lambda</p><p class="medium-feed-link"><a href="https://aws.plainenglish.io/compressing-large-payloads-for-http-apis-in-api-gateway-831d8f63d751?source=rss-a1d710e0dbe5------2">Continue reading on AWS in Plain English »</a></p></div>]]></description>
            <link>https://aws.plainenglish.io/compressing-large-payloads-for-http-apis-in-api-gateway-831d8f63d751?source=rss-a1d710e0dbe5------2</link>
            <guid isPermaLink="false">https://medium.com/p/831d8f63d751</guid>
            <category><![CDATA[women-in-tech]]></category>
            <category><![CDATA[web-development]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[programming]]></category>
            <dc:creator><![CDATA[Amanda Quint]]></dc:creator>
            <pubDate>Mon, 22 Jan 2024 04:16:55 GMT</pubDate>
            <atom:updated>2024-01-22T04:16:55.536Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Reading 82 Books in 2023 — with some commentary]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/illumination/reading-82-books-in-2023-with-some-commentary-20d2621805e3?source=rss-a1d710e0dbe5------2"><img src="https://cdn-images-1.medium.com/max/1238/1*rwWX_Xt3Rpu1wri7VQv2vA.png" width="1238"></a></p><p class="medium-feed-snippet">My Annual Reading Challenge Summed Up</p><p class="medium-feed-link"><a href="https://medium.com/illumination/reading-82-books-in-2023-with-some-commentary-20d2621805e3?source=rss-a1d710e0dbe5------2">Continue reading on ILLUMINATION »</a></p></div>]]></description>
            <link>https://medium.com/illumination/reading-82-books-in-2023-with-some-commentary-20d2621805e3?source=rss-a1d710e0dbe5------2</link>
            <guid isPermaLink="false">https://medium.com/p/20d2621805e3</guid>
            <category><![CDATA[this-happened-to-me]]></category>
            <category><![CDATA[books]]></category>
            <category><![CDATA[self]]></category>
            <category><![CDATA[reading]]></category>
            <category><![CDATA[self-improvement]]></category>
            <dc:creator><![CDATA[Amanda Quint]]></dc:creator>
            <pubDate>Mon, 01 Jan 2024 19:34:42 GMT</pubDate>
            <atom:updated>2024-01-01T19:34:42.716Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Implementing Third-Party HTTP Endpoints in AWS Step Functions Using CDK]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://aws.plainenglish.io/implementing-third-party-http-endpoints-in-aws-step-functions-using-cdk-7b4d3afd8b83?source=rss-a1d710e0dbe5------2"><img src="https://cdn-images-1.medium.com/max/2600/0*q9qCTmBRHz4R-zso" width="3779"></a></p><p class="medium-feed-snippet">Reach out to Third-Party APIs using Infrastructure as Code</p><p class="medium-feed-link"><a href="https://aws.plainenglish.io/implementing-third-party-http-endpoints-in-aws-step-functions-using-cdk-7b4d3afd8b83?source=rss-a1d710e0dbe5------2">Continue reading on AWS in Plain English »</a></p></div>]]></description>
            <link>https://aws.plainenglish.io/implementing-third-party-http-endpoints-in-aws-step-functions-using-cdk-7b4d3afd8b83?source=rss-a1d710e0dbe5------2</link>
            <guid isPermaLink="false">https://medium.com/p/7b4d3afd8b83</guid>
            <category><![CDATA[women-in-tech]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[web-development]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[Amanda Quint]]></dc:creator>
            <pubDate>Thu, 14 Dec 2023 04:17:20 GMT</pubDate>
            <atom:updated>2023-12-14T04:17:20.710Z</atom:updated>
        </item>
    </channel>
</rss>