Senior GPU Software Engineer — CUDA/Python for Warp Adoption

NVIDIA AI; Santa Clara (CA); On-site; USD 184,000 - 357,000; Full time

Benefits offered by this job

Equity

Benefits package

Job summary

NVIDIA Warp team seeks a hands‑on senior engineer to expand Warp adoption for computational physics, AI, and optimization workflows on NVIDIA GPUs. You will own end‑to‑end integration, from native C++ and CUDA changes to production readiness, and work with internal teams to remove barriers to adoption.

You will implement CUDA‑X libraries, expose Python and native APIs, and design new APIs and building blocks to empower researchers and developers across platforms.

Qualifications

• BS/MS/PhD in Computational Science, Engineering, Physics or related field or equivalent experience.
• 8+ years of relevant software development experience in industry, academia, or both.
• Extensive experience building and optimizing CUDA C++ software for NVIDIA GPUs, including debugging and profiling parallel execution and memory usage.
• Experience building computational software used in research or production, with numerical methods or data pipelines.

Responsibilities

• Own and improve the end‑to‑end path for adopting Warp in computational apps, from CUDA changes to production integration.
• Translate CUDA features into Warp functionality, boosting performance and portability across NVIDIA platforms.
• Collaborate with internal teams (ALCHEMI, BioNeMo, PhysicsNeMo, Kit‑CAE) to integrate Warp and support production adoption.
• Select and integrate CUDA‑X libraries and other technologies, exposing them via Python and native APIs.
• Design and implement Warp improvements, adding APIs, abstractions, and reusable building blocks for simulation, optimization, and AI workloads.

Skills

CUDA C++

GPU programming

Performance optimization

Python

Linux

Education

BS/MS/PhD in Computational Science or related field

Tools

CUDA toolkit

CUDA-X libraries

Python APIs

Job description

NVIDIA Warp team seeks a hands‑on senior engineer to expand Warp adoption for computational physics, AI, and optimization workflows on NVIDIA GPUs. You will own end‑to‑end integration, from native C++ and CUDA changes to production readiness, and work with internal teams to remove barriers to adoption.

You will implement CUDA‑X libraries, expose Python and native APIs, and design new APIs and building blocks to empower researchers and developers across platforms.