About this role
Lead and define high-leverage research in AI computing infrastructure, owning the roadmap to connect emerging infrastructure technologies to Lenovo's Hybrid AI strategy. Provide hands-on technical guidance on system architecture, performance, power efficiency, and co-optimization for LLM training, fine-tuning, and inference while partnering with product and engineering to transition research into products.
Skills for this role
System architecturePerformance optimizationPower efficiencyHardware/software co-optimizationLLM trainingLLM fine-tuningLLM inference/servingAI acceleratorsIntelligent computingDatacenter networkingStorage systemsMemory hierarchyKV cache optimizationDistributed trainingDistributed inferenceProductization of researchLLM serving optimizationKV cache managementSpeculative decodingDisaggregated prefill/decodeQuantization-aware servingAgentic AI systemsMulti-agent orchestrationTool-use pipelinesLong-context inferenceMemory-augmented inferenceHybrid AI deploymentCloud deploymentEdge deploymentOn-device NPUs