Sr. Staff Software Development Engineer - Collectives and Network optimization
Core
Drive AMD's strategy, architecture, and tooling to achieve industry-leading AI pre-training and distributed inference performance on AMD GPUs.
Role type
Sr. Staff Software Development Engineer (AI Performance & Optimization)
Builds
High-performance software stack for AI models (LLM, diffusion, multimodal) on AMD GPUs
Domain
AI/ML Infrastructure, High-Performance Computing, GPU Architecture
Deliverable
production ML models
Required skills
Network/NIC/GPU hardware architecture, software optimization, performance modeling, AI frameworks (PyTorch, JAX, vLLM, SGLang), distributed inference, generative model architecture
Preferred skills
Technical leadership, mentoring, cross-functional collaboration, hardware-software co-design
Technologies
ROCm, PyTorch, JAX, vLLM, SGLang, NIC, GPU
Responsibilities
Set strategy and roadmap for Collectives and Network optimizations; Provide guidelines on network load-balancing, workload scheduling, and model sharding; Perform performance tuning, profiling, and analysis of large-scale models; Participate in hardware-software co-design for future optimizations; Develop tools and infrastructure for performance estimation and reporting; Present analysis results to stakeholders and leadership.
Seniority
Sr. Staff, hands-on IC with strategy & mentorship
