Deskripsi Pekerjaan
Synthetix Systems is seeking a Senior AI Infrastructure Engineer to architect, scale, and optimize the distributed computing clusters powering our generative AI foundation models.
In this role, you will bridge the gap between low-level hardware orchestration and high-level model training pipelines, ensuring maximum GPU utilization, zero-downtime training jobs, and rock-solid platform reliability. You will collaborate directly with our research and ML engineering teams in our San Francisco headquarters.
Tanggung Jawab
- Architect, deploy, and maintain large-scale distributed GPU clusters across multi-cloud environments.
- Optimize network topologies, storage fabrics, and CUDA runtimes to maximize training throughput.
- Design resilient orchestration systems using Kubernetes to handle automated failure recovery during model checkpoints.
- Build telemetry, observability, and cost-attribution tooling for compute-intensive workloads.
- Partner with research scientists to eliminate I/O and communication bottlenecks in multi-node training runs.
- Implement zero-trust security and isolation controls for sensitive datasets and proprietary models.
Kualifikasi
- 5+ years of software engineering experience focusing on systems, cloud infrastructure, or platform reliability.
- Demonstrated expertise managing production-scale Kubernetes clusters running distributed ML frameworks (PyTorch, DeepSpeed).
- Proficiency in Go, Python, or Rust for systems-level tooling and automation.
- Hands-on experience orchestrating GPU hardware (NVIDIA H100/A100), InfiniBand networking, and RDMA.
- Deep understanding of high-throughput distributed storage architectures (Lustre, Ceph, S3).
- Proven track record diagnosing Linux kernel, memory, and networking bottlenecks at scale.