GPU Performance Engineer: CUDA Kernel Optimizer

Two Sigma; New York (NY); Hybrid; USD 165,000 - 300,000; Full time

Benefits offered by this job

Medical insurance

Dental insurance

401k match

Life & disability insurance

Hybrid work policy

Tuition reimbursement

Conference sponsorship

Vacation & leave

Wellness activities

Onsite gyms

Job summary

Two Sigma seeks a GPU programming expert to lead the design of GPU-accelerated kernels for financial workloads. You will optimize culture-changing code, manage precision (FP8/FP4), and profile performance with NVIDIA tools across current and next-gen hardware.

You will build reusable GPU libraries and evaluate RAPIDS, CUTLASS, cuBLAS, TensorRT, NCCL, MPI for real-world financial use cases. A strong background in C++/Python and GPU architecture is essential.

Qualifications

• BS or MS in Science, Technology, Engineering or Math.
• Minimum 1 year of experience; 4–10 years preferred.
• Expert-level CUDA programming: kernel development, memory management, stream and graph optimization.
• Deep understanding of GPU architecture: SM structure, warp scheduling, memory hierarchy.
• Experience with performance profiling and optimization of GPU workloads.
• Strong C++ and Python skills, plus familiarity with mixed-precision computation and numerical stability.
• Track record of delivering speedups on real workloads (not just benchmarks).

Responsibilities

• Design and implement GPU-accelerated kernels for financial computation workloads
• Optimize GPU code for throughput, latency, and memory efficiency across current and next-generation hardware (Blackwell, Rubin)
• Develop procedures for precision management (FP8/FP4 training and inference) in financial applications
• Profile and optimize GPU workloads using NVIDIA tooling (Nsight Systems, Nsight Compute)
• Build reusable GPU libraries and abstractions that modeling teams can use without requiring deep CUDA expertise
• Evaluate and integrate GPU-accelerated libraries (RAPIDS, CUTLASS, cuBLAS, TensorRT) for financial use cases

Skills

CUDA programming

C++ & Python

Parallel computing

Performance optimization

GPU architecture

Profiling GPU workloads

Education

BS/MS in science/engineering/math

Tools

Nsight Systems

Nsight Compute

RAPIDS

cuBLAS

TensorRT

CUTLASS

RAPIDS cuDF

NCCL

MPI

Job description

Two Sigma seeks a GPU programming expert to lead the design of GPU-accelerated kernels for financial workloads. You will optimize culture-changing code, manage precision (FP8/FP4), and profile performance with NVIDIA tools across current and next-gen hardware.

You will build reusable GPU libraries and evaluate RAPIDS, CUTLASS, cuBLAS, TensorRT, NCCL, MPI for real-world financial use cases. A strong background in C++/Python and GPU architecture is essential.