主要职责/Your Responsibilities (within 5 lines): Technical key words (at least 3 words): • Test data governance and organization: Participate in the collection, cleansing, annotation, and standardization of test data for cockpit large language models and intelligent agents; assist in building standardized test datasets for cockpit AI scenarios; and perform data quality checks and compliance verification (data security/privacy). • Support for the development and testing of LLMs and Agent. Assist in designing and executing test cases for large models and chatbots; participate in testing and validating the inference performance of large models, as well as testing intent recognition and multi-turn interactions for chatbots in core AI scenarios such as data dashboards, AI automation, and intelligent Q&A; assist the algorithm team in validating performance following model tuning and track the resolution of defects. • Automated Testing and Data Assetization: Develop automated testing scripts for large models and intelligent agents using Python; assist in creating AI testing analysis reports; consolidate testing data assets; and participate in closing the loop on AI testing issues and optimizing the in-cabin AI experience. • Technical Learning and Documentation: Track technological advancements in large models and intelligent agents within the automotive sector; produce AI testing reports and data governance analysis documents. 岗位要求/Required Qualification: • Bachelor’s degree or above in Computer Science and Technology, Data Science, Artificial Intelligence, Vehicle Engineering (Intelligent Connected Vehicle direction) or related majors. • Proficient in Python programming; able to independently complete script development and automated testing tasks; familiar with the fundamental principles of large language models (pre-training/fine-tuning/inference); candidates with relevant coursework, projects, or practical experience are preferred; basic understanding of AI agent technical architectures; interest in in-vehicle AI scenarios. • Basic understanding of core AI functional scenarios in smart cockpits (e.g., voice interaction, navigation, parking, etc.); strong data sensitivity; ability to understand the relationship between data quality and model performance. • Understand the technical architecture and core features of AI agents. Able to conduct scenario-based testing of intent recognition, multi-turn interaction and cross-function collaboration for intelligent cockpit agents. Candidates with verification experience in agent adaptation to in-vehicle scenarios are preferred. • Bonus Points: • Experience with AI/large-scale model-related coursework, lab research, or personal projects; • Familiarity with test data annotation tools and large-scale model fine-tuning tools; • Strong interest in smart cockpits and in-vehicle AI scenarios； • English: CET-4