Principal Architect, Accelerated Computing

at Nvidia

USD 272,000-431,200 per year

SENIOR

✅ On-site

Tech Stack
Tag name is followed by "@" symbol and proficiency level value. About proficiency levels:

1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;

3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;

7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;

10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.

AI API @ 3 Algorithms @ 4 CUDA @ 4 Communication @ 3 Data Structures GPU @ 4 HPC MPI @ 3 Machine Learning OpenCL @ 4 Performance Optimization @ 6 Prioritization @ 6 Profiling @ 4 Python @ 3

Details

NVIDIA is developing software and system architectures for accelerated high-performance computing, scientific computing, machine learning, artificial intelligence, datacenter, and automotive computing. This position offers the opportunity to make a meaningful impact in a fast-moving, technology-focused company.

Responsibilities

Perform in-depth analysis and optimization to ensure the best possible performance on current and next-generation NVIDIA GPUs.
Create and optimize core parallel algorithms, data structures, and reference code for NVIDIA GPUs.
Understand and analyze the interplay between hardware and software architectures, core algorithms, programming models, and applications.
Lead and collaborate with hardware design, software engineering, product, and research teams to guide the direction of accelerated computing.
Work with accelerated computing applications to facilitate software-hardware co-design.
Write and present work through white papers, conference publications, official blog posts, patent applications, and other appropriate materials.

Requirements

Master's degree or Ph.D. in Computer Science, Computer Engineering, or Electrical Engineering, or equivalent experience.
12+ years of relevant work experience.
Strong mathematical fundamentals, including linear algebra and numerical methods.
Passion for performance optimization.
Hands-on experience with massively parallel GPU programming models such as CUDA or OpenCL.
Familiarity with APIs for multi-node communication, such as MPI or OpenSHMEM/NVSHMEM, is a plus.
Strong knowledge of C and C++, with a solid understanding of software design, programming techniques, and algorithms.
Familiarity with threading APIs for multicore CPUs and Unix-style inter-process communication APIs is a plus.
Familiarity with Python is a plus.
Good communication and organizational skills, including a logical approach to problem solving, time management, and task prioritization.
Experience benchmarking, profiling, and characterizing workloads on GPU and CPU clusters.
Creativity, autonomy, and an interest in pushing architectures to their limits.

Compensation and Benefits

Base salary range: USD 272,000–431,250 per year, determined based on location, experience, and the pay of employees in similar positions.
Eligible for equity and benefits.

Applications for this job will be accepted at least until October 24, 2026. This posting is for an existing vacancy. NVIDIA uses AI tools in its recruiting processes and is an equal opportunity employer.