• Develop and advance the inference microservice framework: OpenAI-compatible API endpoints, inference backend integrations (vLLM, SGLang, TensorRT-LLM, Dynamo), middleware, observability instrumentation, and production hardening across cloud, on-prem, and Kubernetes environments • Architect significant new features in open-source codebases, shepherding them through project acceptance and into production • Build and optimize high-performance model download and caching pipelines across multiple cloud storage backends (NGC, HuggingFace, S3, GCS) - parallel transfers, integrity verification, and seamless multi-cloud operability • Implement the model profile and manifest system that ensures NIMs are optimized for every NVIDIA GPU platform - profile selection, validation, and multi-GPU configuration • Develop and refine cloud microservice patterns - service discovery, health checking, graceful degradation, API gateway integration, and end-to-end request lifecycle management - to ensure NIMs operate reliably at scale in diverse cloud deployment environments • Be a role model for high-quality code across Python, Rust, and C/C++, and model guidelines in test-driven development, agentic AI-assisted development, code review, and cross-team collaboration • Mentor teammates and establish high engineering standards for container quality, security, and operability