Pre-Training ML Engineer for Large-Scale Foundation Models

Jump Trading; New York (NY); On-site; USD 300,000 - 350,000; Full time

Benefits offered by this job

Discretionary bonus eligible

Medical, dental, and vision insurance

HSA, FSA, and Dependent Care options

Employer Paid Group Term Life and AD&D

Voluntary Life & AD&D insurance

Paid vacation plus paid holidays

Retirement plan with employer match

Paid parental leave

Wellness Programs

Job summary

Jump Trading Group is seeking an exceptional Pre-Training Research Engineer to lead the development of massive-scale foundation models and the training stack. You will own scalable infrastructure across GPUs/TPUs, design data pipelines, and push efficiency gains that transform how markets are understood and predicted.

You'll work with a team of researchers and ML engineers on cutting-edge techniques, including mixed-precision training and model parallelism, while balancing ambitious research

Qualifications

• Significant, measurable performance improvements in large-scale distributed training (MFU, throughput, convergence, cost-per-token).
• Published research in efficient training methods, scaling laws, architectures, or systems for ML.
• Background in numerical computing, HPC, or distributed systems; familiar with GPUs/TPUs, high-performance networking, Kubernetes/Slurm, OS internals.
• Expertise in Python and deep learning frameworks (PyTorch and/or JAX).
• Advanced degree (MS or PhD) in CS/ML/Physics/Mathematics or equivalent industry experience.

Responsibilities

• Own and drive the entire training stack for foundation models, building fault-tolerant infrastructure across thousands of GPUs/TPUs.
• Engineer data pipelines that stream terabytes per second and optimize kernel performance for training at scale.
• Co-design novel architectures with researchers and push cutting-edge approaches to mixed-precision training and model parallelism.
• Balance ambitious research goals with practical engineering constraints; contribute to live trading improvements.

Skills

Large-scale distributed training

Research publications

Numerical computing HPC

PyTorch/JAX

MS/PhD in CS/ML/Physics/Math

Ambitious research balance

Collaboration & communication

Availability

Education

MS or PhD in Computer Science / Machine Learning / Physics / Mathematics

Tools

Kubernetes

Slurm

Job description

Jump Trading Group is seeking an exceptional Pre-Training Research Engineer to lead the development of massive-scale foundation models and the training stack. You will own scalable infrastructure across GPUs/TPUs, design data pipelines, and push efficiency gains that transform how markets are understood and predicted.

You'll work with a team of researchers and ML engineers on cutting-edge techniques, including mixed-precision training and model parallelism, while balancing ambitious research