ML Inference Engineer - Low-Latency GPU Systems

Susquehanna International Group, LLP; Bala Cynwyd (PA); On-site; USD 110,000 - 150,000; Full time

Job summary

Susquehanna International Group, LLP is seeking a Machine Learning Engineer in Bala Cynwyd, PA. This role focuses on low-latency inference optimization for high-performance model serving systems.

You will collaborate with researchers to optimize performance, evaluate frameworks, and debug GPU memory issues while managing inference workloads effectively. A strong background in modern ML frameworks, programming experience, and understanding of production environments is essential.

Qualifications

• Experience deploying, optimizing machine learning inference workloads in production.
• Programming experience in Python, Java, C#, and systems languages like C or C++.
• Strong understanding of modern ML frameworks like PyTorch.

Responsibilities

• Design and optimize low-latency inference systems for production ML workloads.
• Profile model inference pipelines for performance improvements.
• Debug performance issues related to GPU and CPU coordination.

Skills

Machine Learning inference optimization

Python

C++

PyTorch

Kubernetes

Education

Background in mathematics, physics, or computer science

Tools

CUDA

Triton

GPU clusters

Job description

Susquehanna International Group, LLP is seeking a Machine Learning Engineer in Bala Cynwyd, PA. This role focuses on low-latency inference optimization for high-performance model serving systems.

You will collaborate with researchers to optimize performance, evaluate frameworks, and debug GPU memory issues while managing inference workloads effectively. A strong background in modern ML frameworks, programming experience, and understanding of production environments is essential.