About the Role
You will own infrastructure across data pipelines, GPU clusters, storage and I/O, and model-serving runtime. You will optimize scheduling, throughput, inference latency, cloud infrastructure, infrastructure as code, multi-region deployment, and reliability while making technical design decisions for substantial production systems.
Requirements
•Infrastructure engineering experience
•Cluster scheduling
•GPU
•Data pipeline
•Disk I/O
•Network I/O
•Inference optimization
•Batching
•Quantization
•KV cache
•Cloud computing
•Infrastructure as code
•Multi-region deployment
•Kubernetes
•Python
•Go
•Rust
•C++
•Terraform
•Action model
•World model
•Video understanding
Responsibilities
•Own orchestration and GPU clusters
•Build preprocessing pipelines for raw gameplay video
•Optimize disk and network I/O
•Optimize production inference through batching, quantization, KV cache, and serving runtimes
•Own infrastructure as code, multi-region deployment, and reliability
•Make and carry technical design decisions to production
Benefits
•Equity
•Medical coverage
•Dental coverage
•Vision coverage
•401(k)
•Fertility benefits
•Parental benefits
•Generous PTO
•Daily meals
•Wellness benefits
•Commuter benefits
•Learning and development stipend
