Principal Software Developer - AI Infra Compute

Oracle•Austin, TX

About The Position

OCI (Oracle Cloud Infrastructure) AI Infrastructure is at the forefront of building a cutting-edge, ultra-high-performance GPU platform designed to support AI/ML/HPC workloads. This is your chance to be part of the AI revolution, creating systems that allow customers to scale from tens to thousands of GPUs without compromising performance. Our team is responsible for designing and developing fundamental architectural changes for GPU delivery, health monitoring, triage automation, and diagnostic services. These are essential for running distributed AI/ML/HPC workloads across thousands of GPUs, leveraging technologies like RoCE and Infiniband.

Requirements

BS or MS degree in Computer Science or relevant technical field involving coding or equivalent practical experience
Deep understanding of operating systems, computer networks, and high-performance applications
6+ years’ experience delivering and operating large-scale production systems (1000+ server instances)
Proficient in one programming language (java/python/c/c++/goLang/shell scripting)
Systematic problem-solving approach, strong communication skills, a sense of ownership, and drive.
Proven ability to deliver products and experience with the full software development lifecycle
Self-motivated individuals with a quick learning ability.
Rock-solid developers and distributed systems engineers with a deep understanding of distributed systems and algorithms.
Comfortable diving deep into any part of the stack, as well as software debugging and low-level systems troubleshooting.
Value simplicity and scalability in design and implementation.
Comfortable working in a collaborative, agile environment and eager to learn.

Nice To Haves

Strong background in Linux systems
Familiarity with system-level architecture, data synchronization, fault tolerance, and state management.
General enterprise storage, networking, or computing experience
Experience with Server/GPU hardware architecture and system management.
Experience with Infiniband or RoCE networking
Hands-on experience designing, developing, and operating public cloud service data planes
Good understanding of databases and SQL (MySQL) and caching technologies (Redis, Memcache etc)

Responsibilities

Designing and developing fundamental architectural changes for GPU delivery, health monitoring, triage automation, and diagnostic services.
Running distributed AI/ML/HPC workloads across thousands of GPUs.
Leveraging technologies like RoCE and Infiniband.
Collaborating effectively with various dependencies, including Network and Data Center operations.