AI多模态算法专家-剪映（北京/上海/广州/深圳/杭州）
Core
Design and build multimodal models and LLMs, improve recommendation systems, and apply NLP/multimodal algorithms to understand user editing intent for the Jianying video editing app.
Role type
Senior IC multimodal algorithm engineer (LLM & Recommendation)
Builds
Multimodal pre-trained models, recommendation engines, and intelligent search features for Jianying
Domain
Consumer media / Video editing / AI
Deliverable
production ML models
Required skills
Multimodal LLM pre-training and fine-tuning, RAG, SFT, RLHF, Prompt engineering, Recommendation algorithms (CF, MF, FM, GBDT, DNN, Wide&Deep), Search and retrieval (ElasticSearch, vector databases), Python, C++, Data structures and algorithms
Preferred skills
Top-tier conference publications, Programming/math modeling competition awards, Experience with video editing business requirements
Technologies
LLMs, RAG, SFT, RLHF, ElasticSearch, Vector databases, Python, C++
