Inspiration

Many students struggle to understand why their resume gets rejected even after applying to multiple jobs. Most applicants don’t know the skill gaps between their resume and job descriptions. We wanted to solve this by building an AI system that gives instant, clear feedback like a real ATS (Applicant Tracking System).

What it does

SkillBridge AI analyzes a user’s resume against a job description and:

Calculates an ATS compatibility score Identifies missing or weak skills Provides personalized improvement suggestions Helps users understand how to improve their resume for better job matching

How we built it

We built SkillBridge AI using:

Frontend: HTML + Bootstrap for a simple and clean UI Backend: Flask (Python) for handling requests AI Engine: Google Gemini API for resume-job matching analysis Parser: Python-based PDF text extraction Logic Layer: Custom prompt engineering to generate ATS-style feedback

The system follows a simple flow: Upload Resume → Extract Text → Send to AI → Generate ATS Analysis → Display Results

Challenges we ran into

Handling API rate limits from Gemini during testing Managing file uploads and PDF text extraction errors Structuring prompts to get consistent ATS-style output Fixing Flask template and path-related issues

Accomplishments that we're proud of

Built a working AI-powered ATS system from scratch Successfully integrated resume parsing + AI analysis pipeline Created a clean and responsive UI Implemented real-world job-market use case Made the system stable for demo use despite API limits

What we learned

How to integrate LLM APIs (Gemini) into real applications Importance of prompt engineering for structured outputs Flask routing, template structure, and backend debugging Handling real-world issues like API quotas and failures Building end-to-end AI web applications

What's next for SkillBridge AI

Add downloadable PDF resume reports Build skill-gap visualization dashboard (charts) Add user accounts to track resume improvements over time Improve ATS scoring with hybrid AI + rule-based system Deploy the platform publicly for students and job seekers Add support for multiple resume formats (DOCX, LinkedIn import)

Built With

Share this project:

Updates