Technical Marketing Engineer – Datacenter GPU Clusters & Networking

At AMD, our mission is to build great products that accelerate next‑generation computing experiences—from AI and data centers to PCs, gaming, and embedded systems. Grounded in a culture of innovation and collaboration, we believe real progress comes from bold ideas, human ingenuity, and a shared passion to create something extraordinary. By joining AMD, you’ll discover that our culture is the real differentiator, pushing the limits of innovation to solve the world’s most important challenges while striving for execution excellence, humility, collaboration, and inclusivity.

As a Technical Marketing Engineer (TME) within the Software Product Management organization for AMD’s Data Center GPU Business Unit, you will play a critical role in helping customers achieve optimal performance and efficiency when training and adapting models on AMD Instinct GPUs. Your work will ensure customers can get productive quickly, supported by high‑quality documentation, performance insights, and hands‑on examples that remove friction and accelerate adoption.

Are you an engineering expert with a talent for pushing AI training workloads to the limits? Are you passionate about optimizing large‑scale model training, understanding distributed systems, and analyzing performance bottlenecks across complex environments? We seek someone who thrives at the intersection of performance engineering and technical storytelling—someone who can deeply understand AI training workloads and communicate practical strategies to improve performance, scalability, and efficiency.

• Partner with AMD’s AI software engineering team to develop performance‑focused technical content for AI training workloads, including optimization guides, benchmarking results, scaling studies, and tuning methodologies.
• Serve as a subject‑matter expert on AI training performance across the full model lifecycle, including:
  • Large‑scale pre‑training (foundation models)
  • Fine‑tuning and parameter‑efficient methods (e.g., LoRA, PEFT)
  • Reinforcement learning workflows (e.g., RLHF, RLAIF)
  • Distillation and model compression techniques
  • Quantization‑aware training (QAT)
• Develop and publish deep technical content for training workloads, including:
  • Performance analysis and bottleneck breakdowns
  • Scaling studies (single‑node and multi‑node)
  • Optimization guides for both pre‑training and post‑training workflows
  • Distributed training best practices (data/model/pipeline parallelism)
  • Workload‑specific tuning strategies and competitive positioning insights
• Analyze and optimize training performance across key system dimensions, including compute utilization, memory efficiency, communication overhead, and scaling behavior in distributed environments.
• Engage with internal and external experts to validate performance claims against real‑world scenarios and large‑scale training runs.
• Validate and analyze training performance results from internal benchmarks and customer proof‑of‑concept (POC) engagements, ensuring accuracy, reproducibility, and credibility.
• Partner with engineering to translate low‑level optimizations (kernels, communication patterns, memory usage) into actionable guidance and influence product improvements based on real‑world workload feedback.
• Stay current on industry training frameworks, distributed training strategies, and emerging techniques, identifying where AMD platforms deliver differentiated performance.
• Act as a bridge between engineering, field teams (FAE/FDE), and business stakeholders, ensuring alignment on training performance capabilities, best practices, and messaging.
• Design and deliver technical enablement (e.g., workshops, demos, technical sessions) to help internal teams and customers optimize AI training workloads.
• Build deep…