车机 Ai 算法工程师（大模型 / 多模态 / 端侧部署）

- Adapt and compress Qwen and other LLM VLM models - Build multimodal interaction pipelines - Conduct experiments evaluation and performance analysis - Deploy and optimize multimodal speech and vision AI - Develop on device LLM and VLM models - Document deployment standards and engineering processes - Export convert quantize and integrate inference engines - Improve latency throughput memory and power usage - Optimize CPU GPU NPU inference performance - Optimize KV Cache and operator fusion - Perform model lightening with quantization pruning distillation - Track and validate model compression and acceleration research