GPU Performance Engineer: Low-Latency CUDA Kernels

Priorslab

Bala Cynwyd (PA)

On-site

USD 120,000 - 190,000

Full time

14 days+

Get a reply from this employer

Your resume and cover letter, tailored to exactly what they’re hiring for.

Get past ATS filters

Job summary

Susquehanna seeks a GPU Performance Engineer to build optimized CUDA kernels for low-latency inference. Work with researchers and engineers to understand model structure and translate mathematical ideas into production-grade GPU code.

You will optimize end-to-end inference, profiling performance, and influencing GPU architecture decisions to push latency and throughput. Strong CUDA/C++ skills and a PhD are preferred.

Qualifications

  • Proven ability to design and optimize CUDA kernels for low-latency inference.
  • Strong C/C++ programming and deep understanding of GPU architecture.
  • Experience with memory hierarchy, occupancy, and precision tradeoffs.
  • PhD or equivalent research experience preferred.
  • Exposure to model deployment frameworks and compiler toolchains.

Responsibilities

  • Design, implement, and optimize custom CUDA kernels for latency-critical workloads.
  • Develop fine-grained GPU implementations tailored to model structure.
  • Analyze models to identify bottlenecks and opportunities for parallelization.
  • Collaborate with researchers to translate math into high-performance compute pipelines.
  • Tune kernels, memory layouts, and I/O to improve end-to-end inference.
  • Profile and benchmark GPU performance.
  • Contribute to GPU architecture decisions and best practices.
  • Improve latency and throughput in production inference systems.

Skills

CUDA kernels
C/C++
GPU architecture
memory hierarchy
latency optimization

Education

PhD in a quantitative field

Tools

ONNX Runtime
TensorRT
Triton
TVM

Job description

Susquehanna seeks a GPU Performance Engineer to build optimized CUDA kernels for low-latency inference. Work with researchers and engineers to understand model structure and translate mathematical ideas into production-grade GPU code.

You will optimize end-to-end inference, profiling performance, and influencing GPU architecture decisions to push latency and throughput. Strong CUDA/C++ skills and a PhD are preferred.

Get your free, confidential resume review.

or drag and drop your file here.

Similar jobs

Similar jobs worth comparing

GPU Performance Architect: Low-Latency CUDA Mastery
GPU Performance Architect: Low-Latency CUDA Mastery

SIG Susquehanna • Pennsylvania

On-site
USD 120,000 - 150,000
GPU Performance Engineer | Experienced Hire
GPU Performance Engineer | Experienced Hire

Priorslab • Bala Cynwyd (PA)

On-site
USD 120,000 - 190,000
Latency-Critical CUDA Kernel Engineer
Latency-Critical CUDA Kernel Engineer

Susquehanna International Group, LLP • Bala Cynwyd (PA)

On-site
USD 100,000 - 130,000
GPU Performance Engineer | Experienced Hire
GPU Performance Engineer | Experienced Hire

Trading Interview • Pennsylvania

On-site
USD 140,000 - 210,000
GPU Performance Engineer | Experienced Hire
GPU Performance Engineer | Experienced Hire

SIG Susquehanna • Pennsylvania

On-site
USD 120,000 - 150,000
GPU Performance Engineer | Experienced Hire
GPU Performance Engineer | Experienced Hire

Susquehanna International Group, LLP • Bala Cynwyd (PA)

On-site
USD 100,000 - 130,000
Low-Latency ML Inference Engineer
Low-Latency ML Inference Engineer

Susquehanna International Group, LLP • Bala Cynwyd (PA)

On-site
USD 150,000 - 230,000
GPU Performance Engineer & CUDA Kernel Architect
GPU Performance Engineer & CUDA Kernel Architect

Liquid AI • San Francisco (CA)

On-site
USD 180,000 - 240,000
Equity
Medical/Dental/Vision
401(k) match
+2
GPU Kernel Engineer – AI Inference & CUDA Optimizations
GPU Kernel Engineer – AI Inference & CUDA Optimizations

FriendliAI Inc. • San Francisco (CA)

On-site
USD 180,000 - 240,000
Flexible working hours
Daily meals / snacks provided
Health insurance
+2
Low-Latency ML Inference Engineer (GPU/ Production)
Low-Latency ML Inference Engineer (GPU/ Production)

Priorslab • Bala Cynwyd (PA)

On-site
USD 150,000 - 210,000