AI Research Engineer, Inference

Overview

In this role you will advance large-scale model inference for HRT’s market-facing models within HAIL. You’ll collaborate with researchers to co-design and optimize inference across diverse devices and setups deployed globally, pushing the boundaries of performance and efficiency. The work combines systems engineering with cutting-edge ML, delivering tangible impact on trading outcomes. A strong hook is the chance to shape foundation-model-in-market capabilities at a fast-moving, research-driven trader.

Pay / Benefits

• discretionary performance-based bonuses
• competitive benefits package

Responsibilities

• Improve all aspects of large-scale model inference, including GPU kernel development
• Work on novel inference devices (ASICs, FPGAs) and data streaming
• Co-design with researchers to improve models and shape the research agenda
• Maintain and optimize an inference solution deployed across devices worldwide
• Tackle challenging, high-impact problems in a fast-evolving domain

Key requirements

• Two or more years of experience building deep learning systems in any domain
• Strong engineering skills in CUDA/Triton/Pallas/CuTe DSL kernel development or lower-level PyTorch/JAX/XLA development
• Experience with CUDA Graphs and FPGA/ASIC experience
• LLM experience is valuable but not required
• Finance experience not required

• CUDA/Triton/Pallas/CuTe DSL kernel development
• lower-level PyTorch/JAX/XLA development
• CUDA Graphs
• FPGA/ASIC experience
• deep learning systems engineering

…