Projects
Real problems. Real impact. That’s my approach to projects
(section under construction)
_edited_edited.jpg)
​This was our final year project in UG.
-
Built a predictive model using XGBoost and SVM on a clinical dataset of 50K+ records, achieving 90% accuracy.
Performed hyperparameter tuning, and model interpretability (SHAP values) for actionable insights. -
Unlike most models that only determine illness, our machine learning approach offers both binary and multiple classifications, crucial for early treatment.
-
Developed a sequential CNN model to detect tuberculosis from chest X-rays (radiography) to acquaint myself with bio-medical imaging.
-
As a person who loves both art and tech I found the project interesting as it involved image processing and visualization through AI/ ML.
_edited.jpg)
​
-
India's linguistic diversity creates communication barriers, particularly for business professionals and rural populations.
-
Created a real-time chat app translating 10 Indian languages via langdetect and seq2seq models
langdetect and sequence-to-sequence NLP models -
Built database with sql, backend with Flask, frontend with React, and real-time messaging via WebSockets.
