Director of AI Workload Intelligence
SK hynix America3 days ago
San Jose, CA, USAStaff+
Base Salary
$250k - $300k/yr
Responsibilities
- Lead global collaboration and drive AI workload-based HBF technology development from research through proof of concept and productization.
- Analyze AI models, algorithms, workload trends, and inference data objects for HBF applicability.
- Classify memory objects including weights, KV cache, activations, embeddings, adapters, and MoE experts.
- Define model-level criteria for data placement, caching, prefetching, and offload decisions.
- Analyze HBF integration points in vLLM, SGLang, TensorRT-LLM, PyTorch, and ONNX Runtime.
- Develop memory-use taxonomies and HBF applicability criteria.
- Provide model-driven inputs to Compiler/Runtime and AI Inference Software Stack teams.
Requirements
- Proven experience leading global AI systems architecture initiatives and cross-functional teams.
- Hands-on experience with AI inference systems, ML systems, LLM serving, AI model analysis, AI framework or runtime analysis, or heterogeneous accelerator software.
- Strong understanding of Transformers, LLMs, mixture-of-experts models, recommendation models, embedding and retrieval workloads, and multimodal inference.
- Experience with PyTorch, Hugging Face, vLLM, SGLang, TensorRT-LLM, ONNX Runtime, Triton Inference Server, or equivalent AI inference stacks.
- Experience analyzing latency, throughput, token throughput, memory footprint, bandwidth, and data movement.
Benefits
- Top-tier health insurance at no employee cost.
- Paid time off, company holidays, parental leave, and Happy Fridays.
- 401(k) matching.
- Flexible Spending Accounts for health care and dependent care.
- Educational reimbursement up to $10,000 per year.
- Donation matching and volunteering opportunities.
- Corporate discount programs.
- Free breakfast, lunch, and dinner provided to employees.
- Full-time onsite work in San Jose, California.
Tech Stack
Categories