I am a machine learning engineer focused on graph neural networks, NLP, and production ML deployment, with hands-on experience building efficient GAT architectures and end-to-end inference systems. I design scalable training pipelines, optimize models for latency and memory, and deploy services handling thousands of daily requests with high uptime.
My work includes implementing distributed PyTorch DDP training, INT8 quantization, TorchScript compilation, and automated hyperparameter search with Ray Tune, alongside data engineering and web development projects that reduced no-shows, completed large migrations, and delivered analytics-driven product improvements.