Deskripsi Pekerjaan
Synthetix Systems is seeking an experienced Senior AI Infrastructure Engineer to architect, scale, and optimize high-throughput distributed systems powering our next-generation foundation models. In this role, you will eliminate compute bottlenecks, manage multi-node GPU clusters, and accelerate model training pipelines operating at petabyte scale.
We prioritize zero-overhead architectures, low-latency execution, and resilient distributed storage. If you excel at solving complex systems-level challenges at the intersection of hardware and large-scale deep learning frameworks, this role offers direct impact on our core product suite.
Tanggung Jawab
- Architect and maintain high-performance GPU/accelerator clusters across bare-metal and cloud environments.
- Optimize distributed training frameworks (PyTorch, DeepSpeed, Megatron-LM) to maximize compute utilization.
- Design low-latency data pipelines capable of feeding terabytes of tokenized data per hour to training nodes.
- Implement automated cluster monitoring, health checks, and self-healing node eviction protocols.
- Collaborate with research teams to profile and eliminate kernel-level and networking bottlenecks.
- Enforce rigorous security and isolation controls across multi-tenant inference and training infrastructure.
Kualifikasi
- Bachelor’s or Master’s degree in Computer Science, Systems Engineering, or equivalent practical experience.
- 5+ years of software engineering experience with a primary focus on distributed systems or platform infrastructure.
- Proficiency in Go, C++, or modern Python with direct experience writing production-grade systems software.
- Hands-on experience orchestrating workloads using Kubernetes, Slurm, or Ray on massive GPU clusters.
- Deep understanding of high-throughput networking protocols (InfiniBand, RoCE, NCCL optimization).
- Track record of building automated CI/CD infrastructure and infrastructure-as-code using Terraform.