Senior Lead AI Systems and Performance Architect - LLM Inference

Job Title

Senior Lead AI Systems and Performance Architect - LLM Inference

Role Summary

Lead the technical direction and system-level analysis of an end-to-end LLM inference simulation platform. Drive simulation development, dynamic trace replay, and bottleneck analysis to inform silicon and platform architecture decisions.

Work closely with silicon architects, software engineers, and platform teams; role is based in Shanghai and is eligible for a hybrid work model.

Experience Level

Senior. Typical experience guidance from the posting: Ph.D. with ~4–6+ years, Master’s with ~6–8+ years, or Bachelor’s with ~8–10+ years of relevant experience.

Responsibilities

Deliver and operate a simulation-driven workflow to evaluate full-system LLM inference performance and translate findings into actionable architecture recommendations.

• Own technical direction and continuous evolution of a discrete-event simulation platform modeling distributed multi-accelerator inference systems.
• Model serving behaviors: continuous batching, chunked prefill, prefill-decode disaggregation, and dynamic KV cache allocation.
• Integrate behavioral models for scale-up interconnects (e.g., UALink, PCIe Gen6/7) and capture latency, arbitration, and collective communication effects.
• Design automated simulation pipelines to ingest empirical performance data, run parameter sweeps, and analyze results.
• Drive ingestion and replay of production traces to analyze burstiness, prompt/output distributions, shared prefixes, and Agentic workflows.
• Perform sensitivity and root-cause analyses to identify bottlenecks across compute, memory, interconnect, and scheduling.
• Translate simulation outcomes into silicon and platform recommendations: compute/memory balance, SRAM sizing, interface configurations, and host offload opportunities.
• Mentor engineers on simulation methodology, model calibration, code quality, and experimental rigor; align cross-functional teams on study design and findings.

Requirements

Must-have technical skills and proven experience relevant to system-level LLM inference simulation and performance analysis.

• Hands-on experience developing or using discrete-event, trace-driven, or analytical performance simulators.
• Proven track record of identifying performance bottlenecks across compute, memory, and interconnects in multi-accelerator or heterogeneous environments.
• Strong programming skills in Python and C++ (C++14/17/20) and experience with modular software design and automated data processing.
• Experience ingesting and replaying production traces, including workload characterization for multi-turn conversations and Agentic workflows.
• Ability to design automated workflows for parameter sweeps, empirical table ingestion, and large-scale simulation experiments.
• Experience producing architecture-level recommendations from simulation results and collaborating with silicon and platform teams.
• Technical leadership and mentoring experience; ability to guide team methodology and gain cross-team alignment.

Nice-to-have:

• Familiarity with LLM serving runtimes (examples: vLLM, TensorRT-LLM) and attention/paging mechanisms (e.g., PagedAttention, continuous batching).
• Knowledge of accelerator interconnects and fabrics (CXL, RoCE, UALink, PCIe Gen6/7) and high-performance networking.
• Experience evaluating host-accelerator partitioning, CPU-assisted pipelines, and tiered memory architectures (HBM/LPDDR vs. system memory).

Education Requirements

Ph.D. or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or a related quantitative field. The posting specifies typical experience expectations: Ph.D. with ~4–6+ years; Master’s with ~6–8+ years; Bachelor’s with ~8–10+ years.

About the Company

Company: Intel Corporation

Headquarters: Santa Clara, California, USA

Intel Corporation is a leading multinational technology company known for its innovative semiconductor solutions, including microprocessors, artificial intelligence accelerators, and memory products. Headquartered in the United States, Intel focuses on cutting-edge technology and a collaborative working environment, driving advancements in semiconductor manufacturing to meet global demands. The company emphasizes professional development and aims to shape the future of technology through groundbreaking designs.