- CredX AI - Your Personal Career Trajectory Engine
- π The Big Idea
- π οΈ How We Built It (Hackathon Story)
- βοΈ Run It Yourself
- π How to Use
- π Why CredX AI?
- π‘ Future Improvements
- π₯οΈ Output
- π Live Demo
- π₯ Demo Link
- π€ Contributing
- π License
Ever felt like you're shouting into the void when applying for jobs? You spend hours polishing your resume and sending it out, only to get back a flood of irrelevant listings. We've been there, and we thought: "There has to be a better way."
That's why we built CredX AI β a hackathon project designed to make the job search feel less like a chore and more like a conversation. Instead of just matching keywords, our engine tries to understand who you are and what you really want in your next role.
The problem with most job sites is that they're not very smart. They see "JavaScript" on your resume and show you every single job that mentions it. They don't know if you're a seasoned pro or just starting out, what you want to earn, or what kind of company culture you'd thrive in.
Here's how CredX AI is different:
- It Actually Reads Your Resume β Upload your resume, and our AI instantly gets a sense of your skills and experience to build your profile. No more endless forms.
- You're in the Driver's Seat β You decide what matters most (salary, job title, location, etc.) by adjusting sliders.
- It Understands What You Mean β Knows that a "Lead Product Designer" is relevant to "UI/UX Design" skills, going beyond simple keyword matching.
- It Tells You Why β Every recommendation comes with a short "Match Story", explaining why the role fits you.
We wanted to build something powerful yet practical during the hackathon.
- Backend: Python + Flask (simple and fast for building APIs).
- Frontend: Vanilla HTML, CSS, JavaScript (lightweight and responsive).
- AI Magic:
- Google Gemini 1.5 Flash β Reads resumes, builds profiles, and generates personalized Match Stories.
- Sentence-Transformers β Local lightweight model for semantic matching (e.g., understanding React β Web Development).
- Data: Pandas for handling job listings in-memory.
- Python 3.8+
- pip (Python package installer)
- Google Gemini API Key
credx-ai/
|-- data/
| |-- jobs.csv
|-- static/
| |-- script.js
| |-- style.css
|-- templates/
| |-- index.html
|-- app.py
|-- data_handler.py
|-- matching_engine.py
|-- resume_parser.py
|-- semantic_matcher.py
|-- skills_scorer.py
|-- story_generator.py
|-- pyproject.toml
-
Clone or Create Project Folder
git clone https://github.com/nitinc264/CredX-AI.git cd credx-ai -
Create Virtual Environment
python -m venv venv
Activate it:
- On Windows:
venv\Scripts\activate - On macOS/Linux:
source venv/bin/activate
- On Windows:
-
Install Dependencies
pip install -e .This will install all required dependencies from
pyproject.toml.Note: The first run will download the Sentence-Transformers model (grab a coffee β).
For development dependencies (optional):
pip install -e ".[dev]" -
Add API Key
- Open
app.py - Replace the placeholder:
API_KEY = "YOUR_GEMINI_API_KEY"
- Open
-
Run the Server
python app.py
The server will start at: http://127.0.0.1:5000
- Upload Resume (PDF): AI auto-fills your profile.
- Manual Input: Enter skills, dream job titles, preferences (comma-separated).
- Use sliders under "What Matters Most?" (e.g., salary, job title, location).
- Click "Find My Future Job" to get recommendations.
- A personalized job list appears with short AI-written Match Stories.
- Smarter job matching.
- Personalized insights.
- Faster, easier, and more human job search.
- Real-time job scraping from popular job boards.
- More granular cultural and work-style matching.
- Mobile-first redesign for accessibility.
-
Mehvish @Mehvish1234
Collaborator | Project Mentor
π https://github.com/Mehvish1234 -
Parag Ekbote @ParagEkbote
Collaborator | Project Mentor
π https://github.com/ParagEkbote
We'd love your feedback and contributions! Feel free to fork this repo, open issues, or submit pull requests.
This project was built for a hackathon β free to use for educational and experimental purposes.
