AI Research Engineer, Inference

Hudson River Trading is seeking an AI Research Engineer to join the HAIL team, focusing on large-scale model inference for financial markets. The role involves GPU kernel development, novel inference devices, and data streaming to improve model performance. You will collaborate closely with researchers to co-design models and shape the research agenda.

• Improve all aspects of large-scale model inference
• Develop GPU kernels using CUDA/Triton/Pallas/CuTe
• Work with novel inference devices like ASICs and FPGAs
• Optimize data streaming and inference pipelines
• Collaborate with researchers to co-design models

• Strong engineering skills in CUDA, PyTorch, JAX, XLA
• Experience with CUDA Graphs, FPGA/ASIC
• Two or more years building deep learning systems

• LLM experience
• Experience translating methods between application areas