Senior System Software Engineer - Scientific Computing PaaS

NVIDIA•Santa Clara, CA

3d•$184,000 - $356,500

About The Position

NVIDIA is seeking a Senior System Software Engineer to build scientific computing platform workflows in the cloud. The platform supports physics-based numerical simulation solvers, AI training and inference, and visualization workflows for physical science and engineering applications, including weather prediction, climate modeling, industrial design, and digital twins across domains such as aerospace, automotive, sports, renewable energy, and biomedical science. The role involves designing services and owning cloud infrastructure for physics-informed and data-driven scientific workflows. It also focuses on developing algorithms and improving system performance across application code, microservices, APIs, and heterogeneous CPU- and GPU-accelerated computing. The engineer will design, build, deploy, and operate scalable I/O infrastructure for checkpointing, data loading, and data pre- and post-processing, and optimize compute, storage, and network architecture for physics and simulation-driven applications.

Requirements

BS/MS degree in Computer Science or a related area, or equivalent experience.
10+ years of experience building and operating distributed compute- and data-intensive platform-as-a-service systems in the cloud.
Proven skill in a compiled language such as Go, Rust, or C++.
Strong foundational knowledge of cloud computing, including datacenter architecture, cloud security architecture, CPU, memory and I/O virtualization, resource pooling, and elasticity.
Proven skills in distributed systems and parallel processing, including distributed computation models, topology abstraction, logical time, synchronization, deadlock detection, fault tolerance, failure detection, consensus and agreement protocols, parallel algorithms, shared- and distributed-memory architectures, message passing such as MPI and NCCL, and cluster scalability and performance.
Hands-on debugging skills involving processes, threads, deadlocks and synchronization, scheduling, IPC, memory management, file systems, and I/O structures.
Strong algorithmic thinking and system design skills, including recursion, graphs, trees, stacks, queues, and the design and operation of large-scale, loosely coupled distributed systems.
Self-motivated, with strong interpersonal skills and the ability to work independently across multiple teams with minimal direction.

Nice To Haves

Experience building, deploying, and operating AI platforms on HPC clusters.
Experience building, deploying, and operating cloud-native systems, including distributed storage, scheduling, and orchestration across compute, storage, and network.
Experience configuring and troubleshooting hardware, operating systems, kernels, and compilers for maximum performance.
Hands-on debugging experience optimizing compute, networking, and I/O framework performance.
Extensive experience debugging and customizing third-party source code.

Responsibilities

Design services and take ownership of cloud infrastructure for physics-informed and data-driven scientific workflows.
Design algorithms and work with operations to improve system performance across application code, deep learning frameworks, numerical solvers, microservices, APIs, and heterogeneous CPU- and GPU-accelerated computing.
Design, build, deploy, and operate scalable I/O infrastructure for checkpointing, data loading, and data pre- and post-processing.
Optimize compute, storage, and network architecture for physics- and simulation-driven applications.