When I decided to start learning AWS, I knew I didn’t want to just read documentation—I wanted to build something. I found a great starter project from Tech with Lucy on YouTube:
Build With Me: Automating My Life with AWS & AI

This tutorial walked me through building a simple workflow that automatically processes receipt files, extracts the important data, stores it, and sends me an email summary. It was surprisingly beginner-friendly and helped me get comfortable with multiple AWS services.

What This Project Does

Here’s the general idea: I upload a receipt to the cloud, and AWS takes care of the rest.

  • The file goes into S3 (AWS’s cloud storage).
  • Textract reads the file and pulls out data like the total amount.
  • DynamoDB stores that data in a structured format.
  • SES (Simple Email Service) sends me an email with a receipt summary.
  • Lambda glues everything together and automates the flow.
  • CloudWatch helps monitor logs in case anything goes wrong.

This small project covers a lot of what AWS can do—and it only took me about an hour to set up.

AWS Services I Learned About

Service Purpose
S3 Stores receipt files
Textract Extracts text/data from files
DynamoDB Saves extracted data
SES Sends email with data summary
Lambda Runs automation code (serverless)
CloudWatch Tracks logs for troubleshooting
IAM Roles Controls access and permissions

The Error That Made Me Sweat

Everything was working great until I hit an error when testing the Lambda function:

Log group does not exist

Classic first-time AWS moment.

To troubleshoot, I followed this AWS support article:How do I resolve the "Log group does not exist" error for Lambda function logs in the CloudWatch console?

Here’s what happened:

  • I checked the IAM role attached to the Lambda function.
  • The policy AWSLambdaBasicExecutionRole was attached, but it was missing proper permissions.
  • Specifically, it didn’t allow CreateLogGroup and CreateLogStream.
  • On top of that, the resource name for the log group didn’t match my Lambda function.

Once I corrected the policy and specified the right resource name, everything fell into place. Logs were being generated, the function executed, and emails started landing in my inbox

Plain text email showing successful message of receipts. Three receipts sent to inbox from Amazon SES

What I Learned

This project taught me that AWS isn’t just about spinning up servers—it’s about building tools that can automate parts of your life or work.

Here’s what stood out:

  • AWS Textract is powerful: It can read and pull data from receipts with no manual effort.
  • IAM Roles matter: Even if a policy is attached, it must be properly configured.
  • CloudWatch is a great assist when debugging.

Why You Should Try This

If you're new to cloud computing and AWS, this project gives you:

  • A real-world use case
  • Exposure to 5+ core AWS services
  • The chance to solve an actual error (and learn from it)

It's short, hands-on, and very rewarding.

Give it a go, and let me know how it works out for you!