AURA is a local AI-powered system that generates personalized learning pathways by analyzing resumes against job descriptions — eliminating redundant training and accelerating role readiness.
AURA automates onboarding by identifying skill gaps and recommending a structured learning path.
Inputs:
- Resume (PDF/DOCX)
- Job Description (text)
Outputs:
- Match Score (0–100%)
- Skill Gap Analysis
- Personalized Learning Pathway
- Course Recommendations
- Downloadable PDF Report
- Fully Local ML Pipeline — No external LLM APIs
- Accurate Skill Gap Detection using NLP + fuzzy matching
- Quantified Readiness Score
- Deterministic Learning Pathway (No hallucinations)
- Auto-generated PDF Reports
- Firebase Authentication & History Tracking
- Frontend: React + Vite + TailwindCSS
- Backend: Flask (REST API)
- ML Pipeline: spaCy + Sentence Transformers
- Database: Firebase Firestore
- Auth: Google OAuth
- Extract text from resume
- Identify skills using NLP + taxonomy matching
- Compare with job requirements (fuzzy matching)
- Compute skill gaps and severity
- Recommend courses using semantic similarity
- Generate ordered learning pathway
- React, TailwindCSS, Axios
- Flask, Firebase Admin SDK
- spaCy, Sentence Transformers
- scikit-learn, rapidfuzz
- PyMuPDF, python-docx
- Docker
cd backend
python -m venv venv
venv\Scripts\activate # Windows
pip install -r requirements.txt
python -m spacy download en_core_web_md
python app.py
cd frontend
npm install
npm run dev
docker build -t aura-backend .
docker run -p 5000:5000 aura-backend
Analyzes resume + JD and returns:
- match_score
- gaps
- pathway
- recommendations
Fetch user analysis history
- Corporate onboarding optimization
- Resume skill-gap analysis
- Career roadmap generation
Unlike LLM-based systems, AURA is:
- Deterministic
- Explainable
- Fast (~1–2s)
- Offline-capable
Built for hackathon innovation.
Train people based on what they lack, not what they already know.