职责
——负责电子商务场景中与文字内容（如产品说明、图片文字、口头广播说明等）相关的风险识别模型
-负责产品/短视频/直播评论、报道等舆情内容的风险挖掘算法，提高平台治理效果
——构建行业领先的内容识别方法，构建具有对抗行为的文本内容的跨域、多维识别策略
-探索NLP相关前沿技术，并将其应用于电子商务业务场景
-支持生成可扩展和优化的AI/机器学习（ML）模型
-专注于构建用于提取、转换和加载大量实时非结构化数据的算法，以从理论数据科学模型中部署AI/ML解决方案。
-运行实验来测试部署的模型的性能，并识别和解决过程中出现的错误。
-在团队环境中工作，应用公司要求的统计、脚本和编程语言知识。
-与部署模型的相关软件平台合作。
资格
-扎实的自然语言处理算法基础，对文本分类、相似度匹配、对话问答、机器翻译、序列标注、知识图谱、意图理解、词义消歧等领域有深入的理解和实践经验
——熟悉常用的机器学习和深度学习算法，了解基本的网络模型结构（DNN/LSTM/CNN等）和文本表示方法（LDA/Word2Vec/ELMo/GPT/BERT等），对深度学习训练和推理模型调优有实际经验
-动手能力强，熟练使用至少一个主流深度学习框架（TensorFlow/PyTorch/Caffe/MXNet），了解分布式训练、提炼加速等实现方法
Responsibilities
- Responsible for the risk identification model related to the text content (such as product description, picture text, oral broadcast description, etc.) in the e-commerce scene
- Responsible for product/short video/live comments, reports and other public opinion content risk mining algorithm, improve the platform governance effect
- Build an industry-leading content recognition method, and build a cross-domain and multi-dimensional recognition strategy for text content with adversarial behavior
- Explore NLP-related cutting-edge technologies and apply them to e-commerce business scenarios
- Support the production of scalable and optimised AI/machine learning (ML) models
- Focus on building algorithms for the extraction, transformation and loading of large volumes of realtime, unstructured data to deploy AI/ML solutions from theoretical data science models.
- Run experiments to test the performance of deployed models, and identifies and resolves bugs that arise in the process.
- Work in a team setting and apply knowledge in statistics, scripting and programming languages required by the firm.
- Work with the relevant software platforms in which the models are deployed.
Qualifications
- Solid foundation of NLP algorithm, in-depth understanding and practical experience in text classification, similarity matching, dialogue question and answer, machine translation, sequence tagging, Knowledge Graph, intention understanding, word meaning disambiguation and other fields
- Familiar with commonly used machine learning and deep learning algorithms, understand the basic network model structure (DNN/LSTM/CNN, etc.) and text representation methods (LDA/Word2Vec/ELMo/GPT/BERT, etc.), and have practical experience in deep learning training and reasoning model tuning
- Strong hands-on ability, proficient in using at least one mainstream deep learning framework (TensorFlow/PyTorch/Caffe/MXNet), and understanding of distributed training, distillation acceleration and other implementation methods
- Excellent problem analysis and problem solving skills, have certain processing methods and optimization experience on domain migration, small sample construction, text mining, unsupervised/semi-supervised and other issues
- Master basic big data related components (Hadoop/Spark/Hive/Flink), and have experience in large-scale text data processing and cleaning
- Familiar with Linux development environment, proficient in C++ /go/python at least one programming language, and solid foundation in algorithms and data structures
- Good sense of teamwork and communication skills, practical experience in relevant business scenarios is preferred
- Published papers in high-level computer science conferences (ACL, EMNLP, NIPS, AAAI, etc.) or have competition experience are preferred
发光如星是一家有着独特基因的猎头公司，顾名思义，公司希望每一个发光如星的员工都可以煜煜生辉，同样也希望整个公司能在互联网这个行业中绽放自己的光芒；在其他同业公司都在追求快速扩张的时候，我们在努力的夯实公司发展的底层地基。更关注员工的个人成长和组织的核心稳定性，本着对客户和顾问负责的态度，稳扎稳打，不断打磨自己的培训体系；为行业培养了越来越多的优秀顾问