👋Hi! I am a Machine Learning Engineer with expertise in computer vision and deep learning, currently pursuing MS in Computer Science at Oregon State University. Experienced in implementing CNN and transformer models for industrial applications, with a focus on improving performance for multiclass semantic segmentation tasks with Long-Tail class distribution. Further, I work on optimization and efficient deployment of these models for real-time inference.
- 🎓 MS in Computer Science, Oregon State University (2023-2025)
- 💼 Machine Learning Engineer at Microtec Inc
- 🔬 Graduate Research Assistant at Oregon State University
- 📚 Published researcher in edge computing and deep learning
- Implemented CNN and transformer models for multi-class lumber defect segmentation
- Enhanced model robustness for sensor variations through custom augmentations
- Optimized GPU memory handling and developed custom CUDA kernels
- Developed effective fine-tuning techniques for domain adaptation
- Reduced labeling efforts by 75% while improving model performance
- Developing automated data pipelines for security analysis
- Training graph neural networks for real-time anomaly detection
- Converting system logs to provenance graphs
- Languages: Python, C++, C, JavaScript, Java, SQL
- ML/DL Frameworks: PyTorch, CUDA, ONNX, TensorRT, FFCV
- Libraries: OpenCV, NumPy, pandas, matplotlib, Scikit-learn
- DevOps: Git, Docker, Azure, Jenkins
- Web Technologies: React.js, Node.js
- Implemented 2D/3D UNet, ResUNet, and Attention-ResUNet architectures
- Developed Mixture of Experts architecture improving IOU score by 6%
- Achieved IOU scores of 0.8 (3D ResUNet) and 0.79 (2D Attention-ResUNet)
- Developed and trained ResNet-14 model in PyTorch
- Achieved 83.54% accuracy on CIFAR-10 dataset
- "Inference at the Edge for Complex Deep Learning Applications with Multiple Models and Accelerators" - ICCCNT 2023
- 📧 Email: [email protected]
- 🔗 LinkedIn: Ashwanth Kuppusamy
- 🌐 Location: Corvallis, OR
Feel free to explore my repositories and reach out for collaborations or discussions about machine learning, computer vision, and software engineering!