Responsibilities
1. Participate in the research and development of NLP/CV basic algorithms for TikTok's local life business, deeply understand the needs of local life scenarios (such as merchant information extraction, user evaluation analysis, POI multi-modal content understanding, etc.), and solve practical business problems work closely with the product and operation teams to continuously iterate algorithm effects in the directions of merchant content quality optimization, POI information accuracy, user demand matching, etc., to promote the achievement of business goals 2. Combined with the technical challenges of local life scenes (such as multi-lingual business information processing, cross-modal content association, real-time content understanding, etc.), track cutting-edge research results and promote the implementation of technological innovation, covering but not limited to: multi-modal content understanding, POI information structuring, application of large models in local life scenes, retrieval and correlation optimization and other algorithm fields.
Qualifications
1. Have solid machine learning foundation and mathematical skills, NLP/CV/multimodal/large model/Agent and other related backgrounds are preferred, have independent research and practical business problem solving capabilities, and experience in scenario algorithm implementation (local life, e-commerce, O2O and other fields are preferred) 2. Be familiar with mainstream text/visual pre-training models, and have at least engaged in algorithm development and implementation in any of the following directions: local life information extraction, POI understanding, multi-modal content analysis, large model scenario application, text generation (such as POI Summary generation) proficient in using mainstream machine learning frameworks (such as PyTorch/TensorFlow) and model training tools 3. Have strong technical curiosity and self-motivation, and be able to continuously pay attention to and learn from industry best practices be good at converting business problems into algorithmic models (such as user demand prediction, merchant content quality assessment, etc.), have excellent cross-team communication skills and project driving force, and be able to collaborate efficiently to solve actual business pain points.
