Job Details

About the Company: A technology company focused on model compression, quantization, neural architecture search, LLM lightweighting, and efficient deployment of AI models on specific hardware.

Target Candidates: Senior algorithm experts in AI model compression, inference acceleration, AutoML, LLM lightweighting, and hardware co-optimization

Responsibilities

·       Define the company roadmap for model compression, quantization, pruning, knowledge distillation, neural architecture search, and LLM lightweighting — forming a company-level high-efficiency AI technology plan.

·       Build an automated model-optimization platform and toolchain, improving algorithm R&D, deployment, and iteration efficiency — bridging frontier AI research and productization.

·       Lead the team to overcome algorithm performance bottlenecks on specific hardware, and collaborate with hardware architecture, software, and product teams to define next-generation hardware features and deployment solutions.

·       Manage key customers, partners, and external technical exchanges — supporting improvements in product AI performance and user experience.

Requirements

·       PhD / senior engineer or above, in Computer Science, Electronic Engineering, AI, or related disciplines.

·       5+ years of R&D experience in model compression, inference acceleration, or related fields; 3+ years of algorithm team management.

·       Deep practical experience and successful cases in at least two of model quantization, pruning, knowledge distillation, and NAS.

·       Proficient in PyTorch / TensorFlow and other deep-learning frameworks, with C++ high-performance optimization experience; ASIC / FPGA / accelerator design or hardware co-optimization background is preferred.

Highlights & Support: Compensation negotiable; suitable for overseas AI PhDs or senior algorithm experts with top-conference papers, productization experience, and team management capability.