Microsecond ML Inference Architect

Long Ridge Partners; New York (NY); On-site; USD 600,000 - 1,500,000; Full time

Benefits offered by this job

Generous PTO

Hybrid work options

Free breakfast, lunch, snacks

Job summary

Long Ridge Partners is seeking a Machine Learning Performance Engineer (Inference) in New York to architect low-latency ML inference pipelines for high-frequency trading. You will benchmark CPU, GPU, and FPGA workloads, optimize kernels, and push models into real-time production with strict latency requirements.

You will collaborate with ML researchers, HPC, FPGA, and Datacenter teams to ensure efficient, scalable deployments across hardware platforms.

Qualifications

• 2+ years optimizing deep learning inference in latency-sensitive or high-throughput environments.
• Expertise in lower-level ML framework development (PyTorch/JAX) with solid Python/C++ skills.
• Experience building custom GPU kernels and using optimization libraries (TensorRT, Triton) for low-latency workloads.

Responsibilities

• Lead evaluation of inference platforms (CPU/GPU/FPGA) for latency and throughput.
• Benchmark workloads across architectures and identify performance gains from code vs hardware.
• Develop optimized kernels and integrate libraries to maximize throughput on silicon.
• Design and implement model reduction techniques to meet real-time trading latency targets.
• Collaborate with ML researchers, HPC, FPGA, and Datacenter engineers to productionize deployments.

Skills

Python programming

C++ programming

Mixed-precision compute

Latency optimization

High-throughput systems

Tools

PyTorch

JAX

Triton

TensorRT

ONNX

IREE

HLS4ML

cuBLAS

CUTLASS

Nsight Systems

Nsight Compute

Job description

Long Ridge Partners is seeking a Machine Learning Performance Engineer (Inference) in New York to architect low-latency ML inference pipelines for high-frequency trading. You will benchmark CPU, GPU, and FPGA workloads, optimize kernels, and push models into real-time production with strict latency requirements.

You will collaborate with ML researchers, HPC, FPGA, and Datacenter teams to ensure efficient, scalable deployments across hardware platforms.