I recently had my AWS Community Builder membership extended for another year, and I'm excited to continue building alongside so many talented people in the community.
One of the benefits of the program is AWS credits. As developers, having the freedom to experiment and build without constantly worrying about costs is incredibly valuable.
When I received my credits, I started thinking about what I wanted to build next.
My first thought was Kiro.
Since AWS credits work with Kiro, I set up IAM Identity Center, connected my account, and was ready to go.
Done, right?
Not quite.
The next question immediately became:
"Now what?"
I knew I wanted to build something serverless, but I wasn't sure what.
I started thinking about writing a monthly article summarizing AWS releases and announcements. The challenge was figuring out how to keep up with everything AWS ships.
My first idea was simple:
"Just check the AWS announcements page every day."
Problem solved.
Or so I thought.
Not everyone has time to manually monitor AWS releases every day.
After some research, I discovered AWS RSS feeds that publish release announcements as they happen.
Now things got interesting.
The first version of the project was straightforward:
• Fetch announcements from RSS feeds
• Store them in Amazon S3
• Display them on a webpage
But I wanted more than a feed reader.
I wanted AI to help answer a question I personally care about:
"How does this announcement impact my daily work or personal projects?"
That changed everything.
Instead of simply storing content in S3, I moved toward DynamoDB so I could store structured data alongside each release:
• AI-generated summaries
• Impact analysis
• Categorization
• Additional metadata and insights
Over the course of a day, Kiro helped me build much of the foundation. My role shifted from writing every line of code to making architectural decisions and guiding the implementation.
That said, I still jump into the code regularly. Sometimes I tweak features, optimize workflows, or dig into the implementation to understand what’s happening under the hood. I enjoy coding, and I don’t want to lose that muscle. Tools like Kiro help me move faster, but staying hands-on keeps me sharp as an engineer.
One of the most interesting discussions wasn’t about code at all—it was about cost optimization.
Should the system write a new record every time it processes a release?
Or should it only write when a release doesn’t already exist?
Small decisions like these have a big impact on cost, scalability, and operational efficiency.
What started as "I have AWS credits to spend" evolved into an AI-powered AWS release analysis platform.
And honestly, that's one of my favorite parts of building.
You start with one idea, discover a better one, and keep iterating until the project becomes something you never originally planned.
Here's the evolution of the project from V1 to V2.
What started as a simple RSS-to-webpage pipeline evolved into an AI-powered AWS release analysis platform that helps me focus on the AWS services and topics I care about while also understanding how new releases might impact my work and personal projects.
V1 vs. V2 Architecture
The project is open source, and you can also explore the live version here.
Feel free to clone it, deploy it to your own AWS account, and make it your own.
Whether you want a quick way to stay on top of AWS releases or you're interested in extending the functionality, I hope it provides a useful starting point.
If you have ideas, suggestions, or improvements, I'd love to hear them. Open an issue, submit a PR, or send me a message.
I'm especially curious how others are using AI alongside serverless architectures.
How are you using AI in your side projects today?
Leave me a comment; until next time!
Nad
Top comments (0)