
Weekend Annoying Task Challenge: ResumeTailor AI – Automating ATS Resume Optimization with Amazon Bedrock
Tailoring a resume for every job application is repetitive and time-consuming. I built ResumeTailor AI, an AI-powered web application that automatically analyzes resumes, generates ATS-optimized versions using Amazon Bedrock, and deploys the complete solution using AWS infrastructure that is provisioned automatically with Go.
Vision & What the App Does
Every time I apply for a job, I find myself repeating the same process.
Open my resume.
Read the job description.
Update keywords.
Rewrite bullet points.
Check whether the resume is ATS-friendly.
Save another version.
Repeat for the next company.
It is one of those repetitive tasks that everyone knows is important, yet it consumes a significant amount of time.
For this Weekend Challenge, I wanted to automate that process.
I built ResumeTailor AI, an AI-powered web application that compares a user's resume against a job description and automatically generates an ATS-optimized version in just a few seconds.
Instead of spending 20–30 minutes manually editing a resume for every application, users simply:
- Upload their resume
- Paste the job description
- Click Generate
The application then provides:
- ✅ ATS-Optimized Resume
- ✅ ATS Compatibility Score
- ✅ Missing Skills Analysis
- ✅ Improvement Summary
- ✅ Downloadable Optimized Resume
You can try the live application here:
How I Built It
Although the idea sounds simple, I wanted to build it like a production-ready cloud application instead of a simple AI wrapper.
The project consists of four major parts.
1. Infrastructure Automation
Instead of manually creating AWS resources through the AWS Console, I built a provisioning tool using Go and the AWS SDK for Go v2.
With a single command, the infrastructure tool automatically provisions the required AWS resources.
It creates:
- Amazon EC2 Instance
- Amazon S3 Buckets
- Amazon DynamoDB Table
- IAM Roles
- IAM Instance Profile
- Security Groups
The deployment is idempotent, meaning existing resources are reused instead of duplicated.
This significantly reduces manual setup time and makes deployments repeatable.
2. Backend
The backend is built using FastAPI.
When a user uploads a resume, the backend:
- Validates the uploaded PDF.
- Extracts resume text.
- Uploads the original resume to Amazon S3.
- Sends the resume and job description to Amazon Bedrock.
- Receives structured AI recommendations.
- Generates an ATS-optimized resume.
- Uploads the generated resume back to Amazon S3.
- Stores metadata inside Amazon DynamoDB.
- Returns the ATS score and download link to the frontend.
To improve reliability, I implemented:
- Structured logging
- Request IDs
- Input validation
- Error handling
- PDF validation
- HTTP exception handling
3. Frontend
The frontend is intentionally simple.
The goal of the application is productivity rather than UI complexity.
Users only need to:
- Upload a PDF resume
- Paste a job description
- Generate the optimized resume
- Download the results
Keeping the workflow minimal makes the application fast and easy to use.
4. Deployment
The complete application is deployed on Amazon EC2.
The EC2 instance hosts:
- React Frontend
- FastAPI Backend
- Nginx Reverse Proxy
This allows the entire application to run from a single public endpoint while AWS services handle storage, AI processing, and metadata management.
AWS Services Used
This project integrates multiple AWS services to build a complete cloud-native application.
Amazon Bedrock
Amazon Bedrock is the core AI service powering ResumeTailor AI.
It:
- Compares resumes against job descriptions.
- Generates ATS-friendly content.
- Calculates ATS compatibility scores.
- Identifies missing skills.
- Produces improvement suggestions.
Amazon EC2
Amazon EC2 hosts the complete application.
It serves:
- React Frontend
- FastAPI Backend
- Nginx
The live application is available at:
Amazon S3
Amazon S3 stores:
- Original uploaded resumes
- Generated ATS-optimized resumes
Using S3 keeps application storage scalable and separates file management from application logic.
Amazon DynamoDB
Amazon DynamoDB stores application metadata including:
- Request ID
- Resume location
- Generated resume location
- ATS score
- Processing status
- Timestamp
AWS IAM
IAM Roles and Instance Profiles allow the EC2 instance to securely access AWS services without storing access keys inside the application.
Security Groups
Security Groups protect the deployed application by allowing only the required inbound traffic while keeping the instance secure.
Architecture Overview
The application follows the architecture below.
User │ ▼React Frontend │ ▼FastAPI Backend (EC2) │ ├───────────────┐ │ │ ▼ ▼Amazon Bedrock Amazon S3 │ │ │ ▼ │ Original Resume │ ▼Optimized Resume │ ▼Amazon DynamoDB │ ▼Response Returned to User
Challenges I Faced
This project presented several interesting challenges.
One of the biggest challenges was automating the AWS infrastructure instead of manually provisioning resources.
Building the infrastructure provisioning tool in Go required understanding how different AWS services interact and ensuring deployments could be executed repeatedly without creating duplicate resources.
Another challenge involved integrating Amazon Bedrock while validating AI responses and ensuring structured JSON was returned consistently.
Deploying the complete application on Amazon EC2 also required configuring IAM Roles, networking, storage, and backend services to work together securely.
These challenges provided valuable hands-on experience with real-world AWS development.
What I Learned
This project helped me gain practical experience with several AWS services.
Some of the key takeaways include:
- Building Infrastructure as Code using the AWS SDK for Go.
- Deploying production-ready applications on Amazon EC2.
- Integrating Amazon Bedrock for AI-powered applications.
- Managing file storage with Amazon S3.
- Designing REST APIs with FastAPI.
- Using Amazon DynamoDB for metadata storage.
- Implementing structured logging and production-style error handling.
- Connecting multiple AWS services into a complete application.
Most importantly, I learned that combining multiple AWS services allows relatively small applications to solve real productivity problems effectively.
Project Links
🌐 Live Application
💻 GitHub Repository
Final Thoughts
ResumeTailor AI started as a solution to one repetitive task that almost every job seeker faces—tailoring a resume for every application. What began as a weekend challenge evolved into a complete cloud-native application powered by AWS. Beyond generating ATS-optimized resumes with Amazon Bedrock, the project demonstrates how services like Amazon EC2, Amazon S3, Amazon DynamoDB, and IAM can be combined into a practical, production-style solution, with infrastructure automated through Go.
This challenge pushed me to think beyond simply using AI and instead focus on building a complete application—from infrastructure provisioning and backend development to deployment and user experience. It was an excellent opportunity to strengthen my AWS skills while solving a real productivity problem.
Thank you for reading! If you're preparing for your next job application, feel free to try the app and share your feedback.
Try ResumeTailor AI
Feel free to upload a sample resume, paste a job description, and explore how Amazon Bedrock generates an ATS-optimized resume, compatibility score, and personalized improvement suggestions.
Screenshots:







Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article