Senior software engineer with 15 years of development experience, including 10 years building machine learning frameworks. Custom CUDA kernels and operator optimization; PyTorch/TensorFlow performance tuning (profiling, mixed precision, fusion); distributed training throughput scaling. Also Linux server setup, Docker, and troubleshooting, backed by a home lab I run myself. C++/Java/Python. Evidence-first: profile, measure, optimize — you get numbers, not guesses.... Read more