RoutineIQ

• 

  This picture showcases StudentPathfinder an AI project that predicts exam performance and provide personalized recommendations for students.

About the Project 🌟 Inspiration The inspiration behind this project came from the growing need to understand the rapidly evolving job market. With new roles like AI Prompt Engineer and AR/VR Developer emerging alongside traditional ones such as .NET Developer or Business Analyst, it has become increasingly important to classify and analyze job descriptions efficiently. I wanted to build a solution that not only predicts job roles but also explains the skills and keywords that define them. 🛠️ What We Built We developed a machine learning pipeline that takes job descriptions as input and predicts the most relevant job role or experience level. Using NLP techniques (TF‑IDF vectorization + Logistic Regression), the model learns patterns from text fields like Title, Skills, Responsibilities, and Keywords. Beyond prediction, the project emphasizes #explainability: we extracted the top terms for each role, visualized them, and showed how skills like ASP.NET or GPT directly influence classification. This makes the project useful not only for recruiters but also for job seekers who want to understand what defines a role. 📚 What We Learned

• How to preprocess and clean text data for machine learning.
• The importance of balancing classes in imbalanced datasets.
• How to build explainable AI models that don’t just give predictions but also provide insights.
• Visualization techniques to make results more understandable and impactful. ⚡ Challenges We Faced
• Data imbalance: Some roles had hundreds of examples while others had only a few, making training tricky.
• Text complexity: Job descriptions often contain overlapping skills across roles, which required careful feature engineering.
• #Explainability: Building a model is one thing, but making it transparent and easy to understand was a bigger challenge.
• Time pressure: Hackathons demand quick iteration, so we had to prioritize features that added the most value. 🚀 Motivation & Impact This project is more than just a classifier — it’s a tool to bridge the gap between job seekers and recruiters. By highlighting the most important skills for each role, it empowers individuals to prepare better and helps organizations streamline hiring. The motivation was to create something practical, insightful, and impactful that could stand out in a hackathon setting.

Built With

• colab
• csv
• demo)
• google
• gradio
• matplotlib
• nltk
• numpy
• optional
• pandas
• python
• regex
• scikit?learn
• seaborn
• streamlit