Try here: https://career-copilot-wbkk.onrender.com/
Career Copilot AI is a professional-grade career optimization platform that integrates Large Language Models (LLMs) with cloud-native infrastructure to transform static resumes into strategic career assets.
In a modern job market driven by ATS and precise role-fit, Career Copilot provides job seekers with semantic analysis, predictive interview intelligence, and automated asset generation. It doesn't just scan for keywords; it understands the intent and impact of a user's professional journey, bridging the gap between current experience and future goals.
- Semantic Pathfinding: Uses LLaMA 3.3 to benchmark your resume against 12+ industry-specific roles (AI Engineer, Product Manager, DevOps, etc.).
- Affinity Scoring: A weighted percentage match based on technical skills, soft skills, and leadership experience.
- Skill Gap Mapping: Identifies "Critical Missing Keywords" and provides a prioritized learning roadmap with curated resources.
- Resume-Contextual Questions: AI analyzes your specific projects and tenure to generate the exact questions a recruiter is likely to ask you.
- STAR Framework Guidance: Interactive tips for structuring behavioral answers that emphasize impact and results.
- On-Demand Generation: Use the AJAX-powered engine to generate fresh questions any time you refresh your strategy.
- High-Conversion Cover Letters: Generates context-aware letters that map your unique achievements directly to a job description.
- Career Analysis Reports: Downloadable, high-fidelity PDF reports featuring role-fit charts and detailed strategic advice.
- Live Monitoring: Real-time visibility into user growth, file uploads, and system interactions.
- Audit Logging: A robust internal tracking system that captures logins and system changes for security.
- Feedback Loop: Integrated user feedback system with full CSV export capabilities for administrative review.
- Python 3.12+ – Core backend development and AI orchestration
- JavaScript (ES6+) – Client-side interactivity and dynamic updates
- HTML5 & CSS3 – Web interface structure and styling
- FastAPI – High-performance asynchronous backend framework
- Tailwind CSS – Modern UI design and styling
- Jinja2 – Dynamic HTML templating
- ReportLab – PDF report generation
- Matplotlib – Data visualization and charts
- pdfplumber – Resume text extraction from PDFs
- Gunicorn – Production server for deployment
- Railway – Application deployment and hosting
- Google Cloud Platform (GCP) – Cloud infrastructure services
- Firebase Authentication – Secure Google login
- Cloudinary – Resume storage and file management
- Google Cloud Firestore – NoSQL database for user data and metadata
- ChromaDB – Vector database for semantic search and AI retrieval
- CSV / JSON Storage – Lightweight local persistence
- Groq Cloud API – Fast LLM inference
- LLaMA 3 (70B) – Resume analysis and career guidance
- Firebase Admin SDK – Backend authentication integration
- Cloudinary Upload API – File upload and storage
- Git & GitHub – Version control
- pip – Dependency management
- python-dotenv – Environment variable management
- Mermaid.js – Architecture diagrams
graph TD
subgraph Client_Experience
U(Candidate)
A(Administrator)
end
subgraph API_Orchestration[FastAPI Backend]
Auth[Firebase Middleware]
Parser[PDF Extractor]
LLM[LLaMA 3.3 Engine]
PDF[PDF Report Generator]
end
subgraph Persistence_Layer
FS[(Firestore DB)]
CD[Cloudinary CDN]
SL[(Local Session/Audit Logs)]
end
U -->|Login| Auth
U -->|Upload PDF| Parser
Parser -->|Context| LLM
LLM -->|Scoring| FS
Parser -->|Storage| CD
LLM -->|Data| PDF
PDF -->|Report| U
A -->|Monitor| FS
A -->|Review| SL
sequenceDiagram
participant User
participant FastAPI
participant LLaMA_3.3
participant Firestore
participant Cloudinary
User->>FastAPI: Upload Resume (PDF)
FastAPI->>Cloudinary: Store original file
FastAPI->>FastAPI: Parse text (pdfplumber)
FastAPI->>LLaMA_3.3: Analyze Experience vs Roles
LLA_MA_3.3-->>FastAPI: Return JSON Scoring & Gap Analysis
FastAPI->>Firestore: Cache Results & File Metadata
FastAPI-->>User: Render Interactive Dashboard
- Python 3.12+
- Node/NPM (Optional for Tailwind development)
# Clone the repository
git clone https://github.com/tisha-varma/Career-Copilot.git
cd Career-Copilot
# Create and activate virtual environment
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
# Install dependencies
pip install -r requirements.txtCreate a .env file in the project root:
# AI Intelligence
GROQ_API_KEY=your_groq_cloud_key
# Asset Hosting
CLOUDINARY_URL=your_cloudinary_url
# Identity & Security
ADMIN_EMAILS=user1@gmail.com,user2@gmail.com
SESSIONS_DIR=_sessions
# Firebase Setup
# Ensure firebase-adminsdk.json is present in /app directory- Identity: Sign in with Google to create your personalized profile.
- Analysis: Upload your PDF resume. The system will benchmark you against market standards.
- Strategize: Review your role-fit score and technical gap analysis.
- Prepare: Use the "Interview Prep" section to generate AI-tailored questions for your target role.
- Generate: Create cover letters and download your comprehensive Career Analysis Report.
| Endpoint | Method | Description |
|---|---|---|
/analyze |
POST | Parses resume and generates full career insights |
/download-report |
GET | Generates and exports the Career Analysis PDF |
/generate-resume-questions |
GET | AJAX endpoint for AI question prep |
/submit-feedback |
POST | Submits user ratings for system improvement |
/admin/dashboard |
GET | (Admin) Central monitoring command center |
/admin/delete-user/{uid} |
DELETE | (Admin) Secure profile removal |
Career-Copilot/
├── app/ # Main Application Logic
│ ├── main.py # Central Router & Lifecycle
│ ├── llama_analyzer.py # LLM Analysis Logic
│ ├── firestore_db.py # Firestore Repository
│ ├── cloudinary_storage.py# Asset Hosting Service
│ ├── report_generator.py # ReportLab Orchestration
│ ├── audit.py # Internal System Logging
│ ├── static/ # Visual CSS/JS Assets
│ └── templates/ # Jinja2 Dashboard Layouts
├── data/ # Feedback and Local Sessions
└── requirements.txt # Core Dependencies
- ATS Optimization Score: A dedicated feature to maximize resume pass-through.
- Mock Voice Interviews: Integration with ElevenLabs for roleplay simulations.
- Browser Extensions: Quick analyze features for LinkedIn job postings.
Contributions are welcome! Please follow these steps:
- Fork the Project
- Create your Feature Branch (
git checkout -b feature/AmazingFeature) - Commit your Changes (
git commit -m 'Add some AmazingFeature') - Push to the Branch (
git push origin feature/AmazingFeature) - Open a Pull Request
Distributed under the MIT License. See LICENSE for more information.
Built with ❤️ by Tisha Varma. Helping the world navigate the future of work.