Infrastructure & Pipelines: Build, maintain, and scale the infrastructure, automation, and CI/CD workflows necessary for rapid and efficient model deployment, including large-scale image/video data ingestion, dataset versioning, and managing manual/automated image annotation workflows. - Resource & Cost Optimization: Manage cloud environments and GPU compute resources to ensure systems are not only highly scalable but also cost-effective, focusing on model optimization for heavy GPU workloads, including latency, throughput, batching, and GPU memory utilization.