Terry Kong

Terry Kong is a senior deep learning engineer at NVIDIA, working on model alignment and interested in problems at the intersection of infrastructure and deep learning algorithms. He earned his M.S. in electrical engineering from Stanford University.
Avatar photo

Posts by Terry Kong

Models / Libraries / Frameworks Aug 20, 2025

Reinforcement Learning with NVIDIA NeMo-RL: Megatron-Core Support for Optimized Training Throughput

The initial release of NVIDIA NeMo-RL included training support through PyTorch DTensor (otherwise known as FSDP2). This backend enables native integration with... 7 MIN READ
Generative AI Jul 09, 2025

Reinforcement Learning with NVIDIA NeMo-RL: Reproducing a DeepScaleR Recipe Using GRPO

Reinforcement learning (RL) is the backbone of interactive AI. It is fundamental for teaching agents to reason and learn from human preferences, enabling... 5 MIN READ
Icon image of a chart and search symbol, on a purple background.
Generative AI Dec 17, 2024

Data-Efficient Knowledge Distillation for Supervised Fine-Tuning with NVIDIA NeMo-Aligner

Knowledge distillation is an approach for transferring the knowledge of a much larger teacher model to a smaller student model, ideally yielding a compact,... 5 MIN READ