Pre-Training AI Research Engineer

Wehrtyou; Greater London; On-site; GBP 186,000 - 223,000; Full time

Benefits offered by this job

Performance-based bonuses

Job summary

Hudson River Trading (HRT) is seeking an AI Research Engineer focused on pre-training for the HAIL team. You will work on improving large-scale model training, from kernel development to data loading, parallelism, networking, and fault-tolerance, collaborating with researchers to shape the research agenda.

Our teams operate across multiple modern GPU clusters with a high GPU-to-researcher ratio, and the role offers impactful work with opportunities to influence business outcomes.

Qualifications

• Two or more years of work experience building deep learning systems in any domain.
• Experience translating methods between areas of application is highly valued.
• LLM experience is valuable, but not necessary.
• Finance experience is not required.

Responsibilities

• Improve all aspects of large-scale model training, including kernel development, training data loading, parallelism, networking, and fault-tolerance.
• Collaborate with researchers to co-design and improve models and shape the research agenda.
• Operate within multiple GPU clusters with a high GPU-to-researcher ratio; drive impactful system-level improvements.

Skills

CUDA kernel

PyTorch/JAX/XLA

CUDA Graphs

FPGA/ASIC

Deep learning systems

Tools

PyTorch

JAX

XLA

CUDA

Job description

Hudson River Trading (HRT) is seeking an AI Research Engineer focused on pre-training for the HAIL team. You will work on improving large-scale model training, from kernel development to data loading, parallelism, networking, and fault-tolerance, collaborating with researchers to shape the research agenda.

Our teams operate across multiple modern GPU clusters with a high GPU-to-researcher ratio, and the role offers impactful work with opportunities to influence business outcomes.