Low-Latency ML Inference Engineer

Susquehanna International Group, LLP; Bala Cynwyd (PA); Vor Ort; USD 130.000 - 210.000; Vollzeit

Zusammenfassung

Susquehanna International Group, LLP is seeking a Machine Learning Engineer focused on low-latency inference optimization to help build, tune, and productionize high-performance model serving systems. You will work at the intersection of machine learning, systems engineering, and GPU performance, handling latency-sensitive inference workloads.

You will collaborate with quantitative researchers and engineers to identify bottlenecks, tune runtimes, and deploy scalable solutions, with emphasis on

Qualifikationen

• Proven experience deploying or optimizing ML inference in production.
• Strong Python/Java/C#/systems language experience (C, C++, Rust, Go).
• Familiarity with PyTorch and performance profiling tools.

Aufgaben

• Design, build, and optimize low-latency inference systems for production ML workloads.
• Profile inference pipelines across execution, batching, memory, I/O.
• Evaluate, integrate, and tune inference runtime systems.
• Improve latency, throughput, GPU utilization for production workloads.
• Build benchmarking tools to compare model variants and deployments.
• Debug GPU memory, kernel behavior, CPU/GPU coordination and overhead.
• Collaborate with lower-level systems or GPU experts on optimization.
• Shape model and system design for real latency constraints.

Kenntnisse

ML Inference

Python

Java

C#

C

C++

Rust

Go

Tools

PyTorch

CUDA

Triton

Kubernetes

GPU profiling tools

Jobbeschreibung

Susquehanna International Group, LLP is seeking a Machine Learning Engineer focused on low-latency inference optimization to help build, tune, and productionize high-performance model serving systems. You will work at the intersection of machine learning, systems engineering, and GPU performance, handling latency-sensitive inference workloads.

You will collaborate with quantitative researchers and engineers to identify bottlenecks, tune runtimes, and deploy scalable solutions, with emphasis on