Deskripsi Pekerjaan
Synthetix AI is seeking an exceptional Senior AI Platform Engineer to architect, scale, and optimize the distributed computing infrastructure powering our generative foundation models.
In this role, you will bridge the gap between machine learning engineering and distributed systems reliability. You will design multi-cluster GPU training topologies, streamline large-scale inference pipelines, and accelerate research-to-production cycles across petabyte-scale datasets. If you excel at eliminating distributed I/O bottlenecks and orchestrating elastic GPU clusters, this role is for you.
Tanggung Jawab
- Architect and maintain high-throughput, low-latency GPU infrastructure using Kubernetes and Ray.
- Build automated CI/CD and deployment pipelines for foundation model training and inference.
- Optimize distributed I/O throughput and checkpointing mechanisms across distributed storage layers.
- Develop telemetry, observability, and cost-attribution dashboards for high-density compute workloads.
- Collaborate directly with ML researchers to profile and resolve CUDA memory leaks and communication overhead.
- Harden platform security, access controls, and data boundary protections across multi-cloud environments.
Kualifikasi
- 5+ years of software engineering experience focusing on distributed systems and cloud infrastructure.
- 3+ years operating large-scale Kubernetes clusters in production environments (AWS or GCP).
- Hands-on experience deploying and profiling distributed ML workloads using PyTorch, Ray, or vLLM.
- Proficiency in Python and Go, with deep understanding of Linux kernel fundamentals and networking.
- Demonstrated experience managing GPU resource scheduling, InfiniBand topologies, and NCCL tuning.
- Strong background with Infrastructure as Code tools such as Terraform or OpenTofu.