Software Engineer ML Infrastructure
Posted 2
views
Summary
Builds and scales the GPU infrastructure behind Cursor's AI training: improving training throughput and reliability, planning GPU infrastructure, automating GPU-cluster operations, and building scheduling and data-movement systems for reinforcement-learning workloads, using Python, TypeScript, Rust, Go, Linux, and Kubernetes.
You will improve the throughput and reliability of training, plan and build GPU infrastructure, scale compute environments for reinforcement-learning workloads, automate GPU-cluster operations, and build scheduling and data-movement systems.
Responsibilities
- Collaborate with ML researchers to improve training throughput and reliability
- Plan and build GPU infrastructure
- Improve compute-environment density and scalability for reinforcement-learning workloads
- Automate the building, monitoring, and operation of GPU clusters
- Build workload scheduling and data-movement systems
Requirements
- Systems and infrastructure software engineering
- Python
- TypeScript
- Rust
- Golang
- Distributed storage
- Networking infrastructure
- Linux
- Cloud infrastructure
- Bare-metal infrastructure
- Large-scale systems
- Infrastructure as code
- Configuration management
- Kubernetes
