Responsibilities
Responsible for NLP The application of the language model in the data direction supports model tuning and effect evaluation, and realizes the comprehensive intelligent upgrade and effect guarantee of data products. 1. Provide analysts, operations, decision makers, merchants, etc. with a variety of data applications based on NLP models to improve the efficiency of data decision-making 2. Stimulate the capabilities of NLP models in NL2DSL, docQA and other scenarios to promote business growth 3. Based on business Based on the demand and business data, train and optimize the NLP model to ensure the effect of the model 4. Based on the application scenarios and functional characteristics of the data NLP model product, establish a scientific model evaluation system.
Qualifications
1. Familiar with natural language processing, machine learning and other related technologies, have relevant application experience, and be familiar with the industry-leading NLP language model principles and training Method 2. Understand the specific method of data construction for NLP language model training, and be able to skillfully use machine learning algorithms to clean and structure data 3. Understand distributed training frameworks such as Deepspeed, and have a certain degree of multi-machine multi-card distributed training Training and debugging experience 4. Experience in evaluating the effects of NLP models, or other experience in actual evaluation of search/recommendation effects is preferred 5. Pay attention to the latest cutting-edge technology of international NLP language models, and have practical experience in related open source models 6. Those who have published NLP-related papers in top international conferences/journals are preferred.
