Kinjal Patel

Kinjal Patel is a senior deep learning algorithm engineer on the NVIDIA TensorRT Model Optimizer team. She works on training and inference optimizations leveraging quantization aware training and knowledge distillation to enhance performance and accuracy of large AI models, particularly LLMs.
Avatar photo

Posts by Kinjal Patel

Generative AI

Fine-Tuning gpt-oss for Accuracy and Performance with Quantization Aware Training

Major open-source foundational model releases are an exciting time for the AI community, bringing unique architectural innovations and capabilities. As the... 7 MIN READ