Senior Software Engineer - Capacity
Core
Design and build a centralized, self-serve capacity platform to model demand, forecast requirements, and deliver optimal CPU and GPU capacity across AWS, Azure, and GCP for Snowflake's operations and AI/ML workloads.
Role type
Senior Software Engineer (Infrastructure/Systems)
Builds
Centralized capacity platform for CPU/GPU procurement, reservation lifecycle, demand forecasting, and fleet utilization optimization.
Domain
Cloud Infrastructure, Multi-cloud Resource Management, AI/ML Workload Support
Deliverable
production ML models | infrastructure
Required skills
Large-scale system design, cloud provider compute provisioning (CPU/GPU), capacity planning, system architectural analysis, observability, complex troubleshooting, Go/Python/Java
Preferred skills
Kubernetes at scale, accelerator/GPU fleet management, hardware price/performance analysis, modeling and forecasting, LLMs
Technologies
AWS, Azure, GCP, OpenTelemetry, Prometheus, Kubernetes
Responsibilities
Design and build the capacity platform unifying allocation and procurement; Own the canonical capacity data layer ingesting demand and supply signals; Serve as liaison to Cloud Service Providers for vendor integration; Build planning systems translating demand to hardware requirements; Drive efficiency by instrumenting utilization and recovering stranded capacity; Integrate hardware evolution into the platform; Partner with core services and finance for capacity forecasting; Ensure high availability via on-call rotations.
Seniority
Senior, hands-on IC