Remote Member of Technical Staff (Software Engineer, Inference & Training Platform)

Description:

• Build a self-serve compute platform for training and inference workloads.
• Operate and manage a large GPU fleet across multiple cloud providers.
• Develop scheduling and placement logic to optimize GPU resource usage.
• Support both long-running training jobs and production inference services.
• Manage Kubernetes orchestration for GPU resources across clusters.
• Implement fault tolerance, autoscaling, and observability for the GPU fleet.
• Collaborate with teams to establish a coherent platform architecture.

Requirements:

• Extensive experience with Kubernetes, including custom operators and multi-cluster management.
• Proven experience managing GPU clusters, specifically with NVIDIA hardware and CUDA.
• Familiarity with orchestrating compute across multiple cloud environments (e.g., AWS, GCP).
• Strong understanding of distributed systems, including scheduling and resource allocation.
• Proficient in systems-level programming languages such as Go, Rust, or C++.
• Experience supporting both training jobs and high-availability inference services.
• Ability to take ownership of complex problems and drive solutions independently.

Benefits:

• Opportunity to work with cutting-edge AI and GPU technologies.
• Collaborative environment with a focus on innovation and problem-solving.
• Potential for professional growth and development in a rapidly evolving field.