triton , Jax, Pytorch, Python, Recommender Systems, BigQuery, Deep Learning, Cuda, Spark, Low-latency inference serving, Multimodal Models, Mixed precision, Language Models, Distributed GPU training, Megatron-style frameworks, Data tensor and pipeline parallelism, TensorRT-LLM, DeepSpeed, Triton Inference Server, Large-scale GPU training, vLLM, Model compression, Deep Reinforcement Learning, FSDP

Machine Learning, Data Modeling, Sql Queries, Statistical Analysis, Python Programming

AWS, Python, Unsupervised Learning, Azure, Gcp, Pyspark, Machine Learning, Deep Learning, hyperparameter tuning, model training, Deployment, supervised learning