ML Performance Engineer — Hybrid, High-Impact Training

Tower Research Capital; New York (NY); Hybrid; USD 170,000 - 230,000; Full time

Benefits offered by this job

Generous paid time off policies

Hybrid working opportunities

Savings plans and financial wellness

Job summary

Tower Research Capital is seeking an engineer to bridge quantitative research and high-performance computing, building and optimizing systems used to train machine learning models at scale. You will focus on accelerating data ingestion, distributed execution, kernel performance, and hardware utilization to enable researchers to iterate quickly on complex models.

The role emphasizes end-to-end training efficiency, benchmarking, and collaboration with ML researchers, HPC engineers, and

Qualifications

• 3+ years of experience optimizing machine learning training workloads in high-performance, distributed, or large-scale computing environments.
• Deep knowledge of machine learning frameworks such as PyTorch or JAX, including their execution models, compilation paths, autograd systems, and distributed-training capabilities.
• Strong programming skills in Python and C++, with experience developing or optimizing performance-critical systems.
• Proven experience with GPU kernel development and optimization using technologies such as CUDA, Triton, CUTLASS, cuBLAS, cuDNN, or related libraries.
• Strong understanding of GPU architecture, including streaming multiprocessor execution, warp scheduling, tensor cores, and the memory hierarchy from registers through HBM.
• Experience with distributed-training technologies and communication libraries such as NCCL, FSDP, DeepSpeed, Megatron-LM, XLA, or equivalent systems.
• Proficiency with performance-analysis tools such as Nsight Systems, Nsight Compute, PyTorch Profiler, or comparable tracing and profiling platforms.
• Understanding of high-performance networking, storage, and accelerator interconnects, including technologies such as InfiniBand, RDMA, NVLink, or NVSwitch.
• Demonstrated ability to benchmark heterogeneous compute platforms and make rigorous, data-driven recommendations about performance, scalability, and cost.

Responsibilities

• Training Performance and Benchmarking: Benchmark model-training workloads across CPUs, GPUs, and other accelerator platforms to identify bottlenecks and guide Tower’s compute infrastructure decisions.
• Develop performance models and standardized benchmarks for measuring throughput, utilization, scalability, and time to convergence.
• Distributed Training Optimization: Design and optimize distributed training strategies, including data, tensor, pipeline, and model parallelism.
• Improve communication efficiency across multi-GPU and multi-node environments by optimizing collective operations, topology awareness, and computation–communication overlap.
• End-to-End Training Efficiency: Analyze and improve the full training pipeline, including data loading, preprocessing, memory management, forward and backward passes, optimizer execution, checkpointing, and experiment recovery.
• Identify bottlenecks across compute, memory, storage, networking, and interconnects to increase accelerator utilization and researcher productivity.
• GPU Kernel and Framework Development: Develop and optimize GPU kernels and performance-critical framework components for quantitative machine learning workloads.
• Integrate specialized libraries, compilers, and execution techniques to improve throughput, memory efficiency, and numerical performance.
• Model and Numerical Optimization: Apply techniques such as mixed-precision training, gradient accumulation, activation checkpointing, operator fusion, and memory-efficient optimizers.
• Evaluate tradeoffs among training speed, numerical stability, reproducibility, model quality, and infrastructure cost.
• Training Infrastructure: Partner with HPC and infrastructure teams to optimize workload scheduling, resource allocation, observability, fault tolerance, and reproducibility across shared compute environments.
• Help define the architecture and tooling required to support large-scale experimentation across on-premises and cloud-based infrastructure.
• Cross-Functional Collaboration: Work closely with ML Researchers, Quantitative Researchers, HPC Engineers, Systems Engineers, and hardware specialists to translate research requirements into highly efficient training systems.

Skills

Python

C++

PyTorch

JAX

GPU kernel

CUDA

Triton

NCCL

DeepSpeed

Megatron-LM

Tools

CUDA

Triton

CUTLASS

cuBLAS

cuDNN

Nsight Systems

Nsight Compute

PyTorch Profiler

XLA

Job description

Tower Research Capital is seeking an engineer to bridge quantitative research and high-performance computing, building and optimizing systems used to train machine learning models at scale. You will focus on accelerating data ingestion, distributed execution, kernel performance, and hardware utilization to enable researchers to iterate quickly on complex models.

The role emphasizes end-to-end training efficiency, benchmarking, and collaboration with ML researchers, HPC engineers, and