We are seeking a Sr System Software Engineer to help us build out our scientific computing platform on Nvidia DGX Cloud. We are building a cloud based accelerated scientific computing platform as a service on the Nvidia DGX cloud. This DGX scientific computing cloud platform enables Physics based Numerical Simulation Solvers, AI based Training, Inference and Visualization workflow for physical science and engineering problems.

Those applications include Weather prediction, Climate modeling, Industrial design and Digital twins simulation in various domains e.g Aerospace, Automotive, Sports, Renewable energy, Bio-medical and many more.

Are you passionate about solving rewarding problems at scale? Do you enjoy crafting robust, critical services for compute and data intensive workload? If so, you may be a phenomenal fit for our team!

• Design, Build, Deploy and Operate Cloud native microservices and APIs for scientific computing workload on DGX cloud.
• Design services and take ownership of underlying cloud infrastructure for physics informed and data driven scientific workflows
• Design novel algorithms and actively engaged with operations to increase overall system performance, it spans across the stack e.g. deep understanding of application code e.g DL Framework, Numerical Solvers, Microservices, APIs and Heterogeneous accelerated computing with CPUs and GPUs.
• Design, Build, Deploy and Operate scalable I/O infrastructure for checkpointing, data loading, pre & post processing of data.
• Optimize compute, storage and network architecture specific to physics & simulation driven applications.

• BS/MS degree in Computer Science or related areas or equivalent experience.
• 10+ years experience working on building and operating distributed compute and data intensive platform as a service on cloud
• Proven skill in a compiled language (Go, Rust, C++ or otherwise).
• Strong foundational knowledge in Cloud Computing e.g "The Datacenter is a Computer" architecture, cloud security architecture, virtualization - CPU, Memory and IO, Resource pooling and elasticity.
• Proven skills in Distributed Systems & Parallel Processing e.g System model of distributed computation e.g. topology abstraction, logical time. Synchronization and deadlock detection in distributed systems, Fault Tolerance and Failure Detection, Consensus and Agreement protocols, Parallel algorithms, shared memory and distributed memory architecture, message passing (MPI, NCCL), Cluster scalability and performance.
• Hands on Debugging skills with Process, Threads , Deadlock and Synchronization, Scheduling, IPC, Memory management, File system and I/O structure.
• Strong Evidence on Algorithmic Thinking & System Design skills e.g Recursion, Graph, Tree, Stack and Queue, Large scale loosely coupled distributed system design and operational experience.
• Be self-motivated, have strong interpersonal skills, and be able to work independently with multiple teams with minimal direction.

• Have built , deployed and operated AI platforms on HPC clusters. Have built, deployed and operated cloud native system including distributed storage, scheduling, and orchestration among compute, storage and network
• Configuring and troubleshooting hardware, operating systems, kernel, compilers for maximum performance
• Hands on debugging skills to optimize performance of compute, networking and I/O framework. Extensively worked on third party source code for debugging and customization