3rd Year Data Science Student
Passionate about data analytics, machine learning, and building practical data-driven systems.
Problem: EV and charging station data was scattered and hard to analyze.
Solution: Designed a PostgreSQL star-schema data warehouse, built ETL pipelines, and dashboards for actionable insights.
Tech used: Python for ETL, PostgreSQL for structured storage, Power BI for dashboards — chosen for scalability and visualization.
What broke / Learned: Handling CSV inconsistencies was tricky; learned robust data cleaning and schema design.
GitHub →Problem: Students lacked a unified platform to track study tasks and manage time efficiently.
Solution: Built a web platform with AI-based timetables, Pomodoro timers, task tracking, and water-intake reminders.
Tech used: Python backend for AI scheduling logic, HTML/CSS/JS frontend — easy to prototype and expand features.
What broke / Learned: Syncing real-time tasks was complex; improved async handling and state management.
GitHub →Problem: Filtering spam messages efficiently from raw SMS data.
Solution: Cleaned & preprocessed text data, trained Naïve Bayes and SVM models, and evaluated predictions.
Tech used: Python, Pandas for preprocessing, Scikit-learn for ML models — for speed and reliability.
What broke / Learned: Feature extraction required fine-tuning; learned importance of TF-IDF and cross-validation.
GitHub →Problem: Predicting shipment delays and statuses from historical logistics data.
Solution: Engineered features, trained Decision Tree and Regression models, and evaluated predictions using MAE/RMSE metrics.
Tech used: Python & Scikit-learn for ML, Pandas for data handling — fast experimentation and analysis.
What broke / Learned: Handling missing data was challenging; learned proper feature imputation and evaluation metrics.
GitHub →Problem: Users struggle to improve presentation skills without feedback.
Solution: Built AI system analyzing body language, posture, and eye contact, giving real-time feedback.
Tech used: Python, OpenCV, ML models — real-time detection and feedback with minimal latency.
What broke / Learned: Accurate gaze tracking was tricky; learned calibration techniques and real-time optimization.
Python, SQL, C++, C, C#
Pandas, NumPy, Power BI, PostgreSQL
Scikit-learn, Feature Engineering, Model Evaluation
Git, ETL Pipelines, OpenCV