Sri Vyshnavi K

About Sri Vyshnavi K

Machine Learning Engineer with 3+ years of experience specializing in deep learning…

Experience

1. Machine Learning Engineer

   NVIDIA

   Sep 2024 — Present · CA, US

   Deployed and optimized deep learning models using Python (PyTorch, TensorFlow) and CUDA-accelerated libraries, improving inference speed by 6%, crucial for real-time GPU applications.• Architected scalable ML pipelines with Triton Inference Server, TensorRT, and CUDA, reducing model latency by 9%, enabling faster inference in production.• Optimized model serving on NVIDIA DGX and NGC platforms with Kubernetes, decreasing deployment time by 20% while enhancing GPU utilization efficiency.• Designed and implemented big data pipelines using Apache Spark, RAPIDS, and Dask, improving real-time anomaly detection and predictive analytics accuracy by 12%.• Developed high-performance AI models for computer vision (image segmentation, object detection) and NLP (transformers, BERT-based models), accelerating processing speeds by 15% with TensorRT and cuDNN.• Implemented end-to-end CI/CD pipelines with Git, Docker, and NVIDIA NGC tools, reducing deployment iteration cycles by 12% and ensuring robust version control.• Collaborated with data scientists, MLOps engineers, and developers to integrate optimized ML models into production, ensuring scalable and maintainable AI applications.