Deskripsi Pekerjaan
Nexus Scale Technologies is seeking an experienced Senior AI Infrastructure Engineer to architect, build, and optimize large-scale distributed computing platforms powering next-generation machine learning workloads. In this role, you will design robust orchestration pipelines, accelerate GPU cluster efficiency, and ensure uninterrupted throughput for massive training and inference runs.
We offer competitive equity packages, comprehensive health benefits, flexible remote-hybrid schedules, and modern hardware stipends to support your best work.
Tanggung Jawab
- Architect and maintain high-performance GPU clusters using Kubernetes and bare-metal orchestration tools.
- Optimize network throughput and storage latency across distributed distributed training clusters running multi-node LLM pipelines.
- Design automated CI/CD workflows for rapid deployment and continuous validation of machine learning models.
- Implement telemetry, real-time logging, and alerting systems using Prometheus and Grafana to maximize cluster uptime.
- Partner directly with research teams to profile CUDA workloads, eliminate kernel bottlenecks, and enhance GPU utilization.
- Enforce rigorous security and access controls across production infrastructure and data pipelines.
Kualifikasi
- 5+ years of production experience in infrastructure engineering, site reliability, or DevOps environments.
- 3+ years operating large-scale distributed systems and container orchestration platforms (Kubernetes).
- Demonstrated proficiency in Python, Go, or Rust for systems-level automation and tooling.
- Practical experience configuring and troubleshooting NVIDIA GPU compute pipelines (CUDA, NCCL, InfiniBand).
- Deep understanding of Linux kernel internals, distributed storage systems (Ceph, Lustre), and networking protocols.
- Track record with Infrastructure as Code tooling (Terraform, Pulumi) in AWS, GCP, or bare-metal setups.