AI-Powered Career Intelligence System
An advanced, all-in-one monolithic application designed to help students and professionals navigate their career paths using intelligent predictions, automated resume parsing, and actionable AI-driven guidance.
โ Project Objective
This project aims to bridge the gap between academic performance and industry expectations. It provides a personalized approach to career planning by analyzing resumes and academic metrics, and offering a Generative AI based Chatbot for customized career support.
๐ง Features & Moduels
1. ๐ Academic Profile & Prediction
A machine learning module built using scikit-learn (Random Forest) that calculates a user's probability of securing a placement based on quantitative (CGPA, marks) and qualitative (internships, extracurriculars) factors.
2. ๐ Resume Analyzer
A Natural Language Processing (NLP) tool that processes uploaded PDF resumes using PyPDF2. It cross-references extracted text against industry standard skill requirements for roles like Machine Learning, Data Science, and Software Engineering, giving the user a proficiency score.
3. ๐ก Intelligent Recommendations
A deterministic rules engine evaluates the placement probability and resume score to produce dynamic, concrete guidance on how to prioritize next steps.
4. ๐ฌ Career RAG Chatbot
An entirely local generative AI assistant utilizing Retrieval-Augmented Generation (RAG). Powered by google/flan-t5-small and faiss-cpu, it strictly references a custom knowledge base to accurately answer questions regarding career transitions and interviews.
๐ ๏ธ Tools & Technologies Used
- Frontend & UI: Streamlit
- Visualizations: Plotly Express
- Machine Learning: scikit-learn, Pandas, NumPy
- NLP & Document Parsing: PyPDF2
- Generative AI & Search: Transformers (HuggingFace), Sentence-Transformers, FAISS, PyTorch
๐งช Try It Live
Click here to access the Career Intelligence System Live Interface
๐ฆ GitHub Repository
GitHub Repository - Career Intelligence System
๐ Connect with Me
๐ www.tauqueeralam.com
๐ฑ LinkedIn | GitHub