A showcase of engineering and machine learning systems spanning Edge AI & Computer Vision, Time-Series Forecasting, Full-Stack Web Architectures, and Hardware Logic Design.
Published:
Resource-efficient 3D-CNN architecture containing only 486K parameters (1.86 MB model size) achieving 88.7% accuracy and 7.19 ms latency on edge devices with spatiotemporal Grad-CAM attribution. Read more
Published:
Multi-tier hybrid ensemble combining gradient-boosted decision trees with regularized meta-regressors, achieving 0.0006 test RMSE with policy-grade SHAP/LIME interpretability on high-frequency indicators. Read more
Published:
Full-stack educational and skill-sharing web platform built with Next.js, React, Node.js, and MongoDB featuring JWT authentication, role dashboards, and RESTful APIs. Read more
Published:
Empirical time-series forecasting evaluating Long Short-Term Memory (LSTM) recurrent neural networks against classical ARIMA statistical models on historical equity data. Read more
Published:
Autonomous vehicle engineered from fundamental discrete logic gates, IR optical sensors, comparator ICs, and H-bridge motor drivers without microcontrollers. Read more