Transforming Pixels into Insights & Silicon into Intelligence Passionate about building high-performance deep learning systems, optimizing CV pipelines, and exploring the intersection of AI with Computer Architecture.
- Deep Learning: CNNs, Transformers (ViT), and Generative Models.
- Computer Vision: Object Detection (YOLO/Faster R-CNN), Segmentation, and Tracking.
- Edge AI: Deploying models on RISC-V and ARM using OpenMP for performance optimization.
- Goal: Designing efficient AI that runs seamlessly on low-power hardware.
| Category | Skills & Frameworks |
|---|---|
| Deep Learning | |
| Computer Vision | |
| Data & Deployment | |
| System Programming |
Custom YOLOv8 pipeline for tracking pucks and players on ice.
- Optimization: Achieved 30+ FPS using TensorRT on Edge devices.
- Impact: Developed as part of a Mitacs Research application for sports performance analysis.
U-Net implementation for identifying anomalies in MRI scans.
- Tech: PyTorch, Albumentations, Dice Loss.
- Result: 94% mIoU on the BraTS dataset.
Multi-spectral image analysis for environmental monitoring.
- Tech: TensorFlow, Earth Engine API, CNNs.
"In the world of pixels, I seek the ground truth."