Megh Makwana

Megh Makwana is an applied deep learning solution architect and engineering manager at NVIDIA. He technically leads Sovereign AI initiatives in India where he works with Sovereign Foundational Model Builders, NVIDIA Cloud Partners, Government, and Enterprise. He specializes in bringing in best practices around large scale data curation, LLM pretraining, post-training, inference optimization, and deployment, and ensuring organizations can achieve maximum efficiency on their GPU infrastructure. He serves as subject matter expert on NVIDIA platforms like NeMo, Megatron-LM, TensorRT-LLM/vLLM/SGLang, DGX Cloud Lepton and NVIDIA Cloud Functions.