As a senior member of the LLM inference framework team at AMD, you will be responsible for building and optimizing production-grade inference runtimes for large language models on AMD GPUs. This role focuses on enhancing performance, scalability, and reliability through various parallelism techniques. You will work closely with inference engines, distributed systems, and GPU runtime backends. Your contributions will directly impact customer-facing deployments and open-source inference frameworks.

• Master's or PhD in Computer Science, Computer Engineering, Electrical Engineering, or a related field.
• Hands-on understanding of vLLM, SGLang, or similar inference stacks.
• Strong experience integrating optimized GPU performance into machine-learning frameworks such as PyTorch or TensorFlow.
• Expertise in Python and preferably experience in C/C++, including debugging and performance tuning.
• Experience running large-scale workloads on heterogeneous GPU clusters.

• Architect and optimize distributed LLM inference runtimes based on in-house LLM engines or open-source stacks.
• Design and improve hybrid execution techniques including KV-cache management and token scheduling.
• Implement and optimize multi-node inference pipelines using RCCL, RDMA, and collective-based execution.
• Drive throughput, latency, and memory efficiency across single-GPU and multi-GPU clusters.
• Collaborate with compiler teams to unblock framework-level performance and ensure efficient use of GPU libraries.

• AMD provides a competitive 'Total Rewards' package that focuses on financial growth, health, and work-life balance.