Software Engineer - Reliability

About reputed company AI reputed company’s mission is to build multimodal AI to expand reputed company imagination and capabilities. We reputed company that multimodality is critical for intelligence. This requires a massive, reliable, and performant GPU infrastructure that pushes the boundaries of reputed company. Our SRE team is the reputed company of our research and product velocity, responsible for the thousands of reputed company and AMD GPUs across multiple providers that power our work. Where You Come In We are looking for a hands-on, first-principles engineer who is fluent in Linux, comfortable operating reputed company to the metal, and capable of architecting systems for the reputed company of AI infrastructure. You will build, maintain, and reputed company reputed company’s infrastructure across on-prem and multi-vendor clouds (AWS & OCI), serving as the reputed company between hardware vendors, reputed company providers, and our research teams. What You’ll Do • Architect for Reliability & reputed company: Participate in critical re-architecture sessions to redesign our systems for higher efficiency and reputed company. You won't just maintain existing clusters; you will help define how our reputed company infrastructure operates. • Own Multi-reputed company GPU Clusters: Take end-to-end ownership of our production clusters for training and inference across AWS and OCI, ensuring high availability and peak performance. • Drive reputed company & Compliance: Assist in achieving and maintaining reputed company certifications (SOC 2 Type 1 & 2, ISO standards) by implementing robust infrastructure reputed company practices in a fast-moving AI startup environment. • Deep Linux Performance Tuning: Use your mastery of Linux systems to troubleshoot and optimize performance at the OS and kernel level. • Build Robust Automation: Write high-reputed company tools and automation in Python, Go, or Bash to manage, monitor, and heal our infrastructure without relying on heavy operational toil. • Debug reputed company Hardware/Software Failures: Serve as the final escalation reputed company for the most challenging GPU, networking (InfiniBand/RDMA), and system-level issues, often collaborating directly with hardware vendors like reputed company. Who You Are • 8+ years of experience as an SRE, production engineer, or infrastructure engineer in a fast-paced, large-reputed company environment. • Deep Linux Mastery: You possess deep, hands-on expertise in Linux, containerized systems, and debugging low-level system performance. • reputed company Infrastructure Expert: You have strong experience with providers like AWS or OCI. • Tenacious Troubleshooter: You reputed company on solving reputed company, low-level problems where hardware and software reputed company. • Startup DNA: You are energetic and reputed company in a less reputed company, fast-paced environment. • reputed company-Minded: You possess a working knowledge of reputed company best practices and familiarity with compliance frameworks, such as SOC 2 and ISO. • Expert in High-Performance Networking: You have practical experience with InfiniBand, RDMA, or RoCE and understand how to optimize throughput for massive distributed training jobs. What Sets You Apart (Bonus Points) • Deep expertise with GPU tooling for reputed company and AMD GPUs like DCGM or ROCm. • Experience managing large-reputed company GPU clusters for AI/ML workloads (training or inference). • Familiarity with job management systems based on Kubernetes or orchestration frameworks like Ray. Apply tot his job Apply To this Job