How I Built an AWS Project Idea Generator Using AI

As someone who constantly learns and writes about AWS, I’ve found myself often asking:

“What should I build next?”

Not just for the sake of learning—but to challenge myself, grow professionally, and create blog posts that actually help others.

But here’s the thing… coming up with project ideas is hard.

🧭 The Problem: Too Many Tools, No Direction

With over 200 AWS services, it’s easy to feel overwhelmed.

I’d often sit down thinking:

• Should I build something with Lambda?
• Maybe try a serverless ETL pipeline?
• What’s a good way to combine AI and AWS?

But instead of starting, I’d spiral into:

Paralysis by analysis.

It turns out I wasn’t alone. Many AWS learners and professionals face the same challenge. That’s when an idea struck:

💡 What If AWS Itself Could Tell Me What to Build?

What if I could describe:

• My experience level
• The AWS services I want to learn or showcase

...and an AI could suggest a tailor-made project idea complete with:

✅ A real-world problem
✅ Architecture with selected services
✅ Career relevance
✅ AWS Well-Architected Pillars in action

That’s when the AWS Project Idea Generator was born.

🧠 The Vision

I wanted it to feel smart but simple.

🗨️ “I’m an intermediate AWS user and want to work with Lambda, DynamoDB, and Step Functions.”

And the generator would reply with something like:

🧩 Problem Statement: Create a system to automatically back up data from DynamoDB tables and restore it when needed.

🧰 AWS Services Used: Lambda, DynamoDB, Step Functions

🛠️ Step-by-Step Plan: Create a Lambda function to export data from DynamoDB tables. → Schedule the Lambda function to run periodically using AWS Lambda's event triggers → Store the exported data in another DynamoDB table or a backup service. → Use Step Functions to orchestrate the backup and restore processes. → Implement a restore process using another Lambda function triggered by Step Functions

🧑‍💼 Resume/Career Relevance: Demonstrates understanding of data management, automation, and orchestration.

🧱 Well-Architected Pillars: Operational Excellence

Now that is a project idea I’d actually build.

🚀 The Spark: Amazon Bedrock

When Amazon Bedrock was launched, I was excited to try powerful foundation models like Titan, Claude, without having to manage infrastructure or train anything.

So I used Bedrock to analyze the user’s selections and generate personalized, full AWS project ideas. This tool wasn’t just an AI gimmick—it became my own career mentor.

🛠️ How It Works

Here’s how simple it is:

1. 🎯 Select your skill level – Beginner, Intermediate, or Expert
2. 🧰 Choose AWS services you want to work with

3. 🤖 AI generates a project idea with:

   • Problem Statement
   • AWS Services Used
   • Step-by-Step Plan
   • Resume/Career Relevance
   • AWS Well-Architected Pillars addressed

4. 💾 Try it live: 👉 Launch the AWS Project Idea Generator

5. 🖼️ See it in action:

🧪 What I Learned

This wasn’t just about building a tool—it was about building confidence.

Along the way, I:

• ✅ Used Amazon Bedrock for tailored AI generation
• ✅ Built a serverless backend using Lambda + API Gateway
• ✅ Thought deeply about career-focused learning paths in AWS
• ✅ Realized AI can be a mentor, not just a chatbot

🙌 Why You Should Try Building Something Like This

If you’re wondering:

• 🤔 What AWS project should I build to grow?
• 📄 What can I showcase in my resume?
• 🧠 How do I apply the services I’ve learned?

Then build something personal—that solves your own roadblocks.

This project solved mine: not knowing what to build next.

Now, I don’t just sit and wonder.

I ask my app—and it answers like a mentor.

🧵 Final Thoughts

The cloud is massive. AI is everywhere.

But the real magic happens when you use both to solve a simple, personal problem.

That’s where growth begins.

Start there.

💬 Feedback is Welcome!

This project was built with a backend-first mindset — focusing deeply on logic, architecture, and AWS integration.

UI/UX might not be perfect (I’m still learning there!), but the goal was functionality + value.

If you’ve got:
💡 Suggestions to improve
🧠 Ideas to extend the tool
🐛 Found any bugs
🎨 UI feedback (go easy on a backend guy 😄)

I'd truly appreciate it! Your input helps me learn and build better.

✍️ Written with ❤️ from the cloud

— Utkarsh Rastogi | AWS Community Builder

🔗 Also writing at awslearner.hashnode.dev

🔗 Connect on LinkedIn