1. Proficient in natural language processing, including word segmentation, part-of-speech tagging, named entity recognition, syntactic parser, question and answer system, intent recognition, sentiment analysis, etc.; 2. Proficient in machine learning and deep learning, such as XGBoost, Hidden Markov Model (HMM), Conditional Random Field (CRF), CNN, RNN, LSTM, Transformer, BERT, GPT pre-training model, etc.; 3. Be familiar with various natural language open source frameworks, such as Sequence_tagging, Transformers, HanLP, etc., and be able to modify the source code based on needs; 4. Familiar with natural language annotation systems and corpora, such as: University of Pennsylvania Treebank, People's Daily Corpus, etc.; 5. Proficient in the architecture and development of similar recommendations and real-time personalized recommendation algorithms based on tags and user behavior;