大模型应用算法Leader-AI应用与创新
Core
Lead multi-modal algorithm research and development for Douyin's ecosystem, focusing on content understanding, user feedback analysis, behavior sequence modeling, and large model evaluation.
Role type
Senior IC multi-modal algorithm leader (LLM/MLLM)
Builds
Multi-modal content understanding features, user feedback systems, and large model evaluation tools for Douyin products
Domain
Social media, video streaming, recommendation systems
Deliverable
production ML models
Required skills
NLP, computer vision, multi-modal learning, large language models (LLM), multi-modal large language models (MLLM), reinforcement learning, pre-trained models (BERT/Transformer/GPT), TensorFlow, PyTorch, Keras
Preferred skills
Large model base research experience, GPU inference acceleration optimization, publications in top-tier conferences (NeurIPS/ICLR/ICML/CVPR)
Responsibilities
Develop multi-modal algorithms for various business scenarios, track latest large model research to improve application effects, collaborate with product and engineering teams to drive project efficiency and revenue
Seniority
Senior, hands-on IC with team leadership
