At MatrixFit Trainings, I automated personal-training session tracking and management notifications, reducing late payments by 80%. I also created a daily task sheet for AM and PM management handovers, cutting average task-completion delays by 50%.
For my final-year project, I developed a Python machine-learning pipeline to predict BRCA1/BRCA2 mutation pathogenicity using more than 33,000 ClinVar records and sequence and structural features. A combined Random Forest model achieved 89.2% peak accuracy and a 0.931 ROC-AUC, and I used SHAP to interpret its predictions.
I’ve also worked on business intelligence projects analyzing eBay car listings and healthcare insurance claims, and built an AI voice receptionist and booking system using Vapi.ai, Supabase, and Node.js. As a Quality Assurance Intern at Amploy, I tested website features, documented bugs, and verified fixes.
