Deskripsi Pekerjaan
Synthetix Core Labs is seeking an experienced Senior AI Infrastructure Engineer to design, scale, and optimize our distributed GPU training clusters and inference backbones.
In this role, you will collaborate with machine learning researchers and platform engineers to build fault-tolerant compute infrastructure capable of running petabyte-scale training workloads and low-latency inference services. We offer top-tier compensation, significant equity, and direct ownership over critical compute systems.
Tanggung Jawab
- Architect and maintain distributed GPU clusters across multi-cloud environments.
- Automate training orchestration pipelines using Kubernetes and Slurm.
- Optimize network topologies for high-throughput, low-latency distributed model training.
- Design zero-downtime deployment pipelines for large language model inference APIs.
- Implement comprehensive observability, distributed tracing, and alerting for mission-critical clusters.
- Diagnose and resolve performance bottlenecks in high-performance distributed storage systems.
Kualifikasi
- 5+ years of experience managing production infrastructure on Linux systems.
- 3+ years operating large-scale Kubernetes clusters in AWS or GCP environments.
- Hands-on experience with GPU workload scheduling and distributed training frameworks.
- Proficiency in Go, Python, and Infrastructure as Code tooling such as Terraform.
- Strong understanding of high-performance networking including RoCE and InfiniBand.
- Proven background diagnosing kernel-level and distributed systems anomalies under pressure.