Algorithm - AI System Researcher

Highlights

• Research next-gen NPU serving to push the roofline of inference performance.
• Explore cluster-level serving and heterogeneous systems via modeling and simulation.
• Validate key hypotheses on real NPU hardware through POCs and prototypes.

Ideal candidate

A curious systems thinker who thrives in ambiguity and loves structuring open-ended problems into testable hypotheses, driven by pushing inference efficiency through analysis and real-hardware validation.

Requirements

• Master's degree (or expected) in AI/ML, Computer Architecture or related field or equivalent experience
• Understanding of LLM inference principles and performance characteristics including attention, KV cache, prefill/decode, batching and multi-chip parallelism
• Ability to perform performance modeling, simulation and data-driven analysis using Python, PyTorch and related tools
• Ability to structure ambiguous undefined problems and drive hypothesis-validation cycles independently
• Ability to communicate analysis results and technical judgment through clear documentation and presentations
• Experience analyzing and optimizing distributed systems, cluster-level software or large-scale workloads
• Understanding of LLM serving system internals such as vLLM, SGLang or TensorRT-LLM
• Experience with quantitative system evaluation such as roofline analysis, analytical performance models or TCO analysis
• Experience optimizing or developing accelerator-based systems such as GPU/NPU
• Familiarity with research and development using AI coding tools