Responsibilities
1. Business application: Responsible for applying self-developed algorithm models to enterprise collaboration software, creating the best AI tools and products in many office scenarios such as meetings, documents, messages, office agents, etc., and continuously improving user experience 2. Model optimization: Responsible for training large language models to continuously improve the algorithm quality in the office field building efficient evaluation methods and technical systems collecting, researching and producing high-quality data sets in the office field 3. Technology construction: continuing to pay attention to new technology trends and research results in the industry, sharing industry best practices, and applying cutting-edge technologies to large models.
Qualifications
1. Have relevant experience in NLP, LLM application or RAG, one of which is enough 2. Strong programming ability, proficient in programming languages such as Python relatively familiar with AI-related Python libraries, such as Pandas, etc. 3. Strong innovation ability, with a strong willingness and subjective initiative to continuously explore new AI technologies and application scenarios 4. Keep up with new developments in the field of AI and continuously improve your professional knowledge and skills by participating in academic conferences, reading cutting-edge papers, online learning courses, etc. 5. Have strong problem-solving abilities and be good at solving various tedious technical problems such as data quality issues, training issues, deployment compatibility issues, etc. Be sensitive to data and be more meticulous in data work, including data quality, data distribution, case analysis, data tuning, etc. Bonus points: 1) Familiar with the concepts of traditional machine learning, familiar with the definitions, goals, problems solved, and measurement indicators of various machine learning tasks 2) Familiar with deep learning frameworks, such as TensorFlow, PyTorch, etc. you can use these frameworks to quickly build and train models to solve practical problems, such as using PyTorch to build a text classification model in natural language processing 3) Have a good foundation in mathematics, understand probability theory, linear algebra, and calculus, be able to analyze the performance and errors of models through mathematical methods, and understand learning tasks expressed in formula definitions.
