- 地点
- 上海，中国
- 工作方式
- 现场办公
- 类型
- 实习
- 发布
- 2026-04-03
- 部门
- 内部/合作
- 编号
- 26004009
职责
- 支持在内部“智能”用例（如知识搜索、内容撰写和生产助手）中部署、优化和改进大型语言和多模态模型。
- 将大型模型 API 打包并集成到内部应用程序和业务工作流程中，以实现跨功能的智能升级。
- 帮助组织和构建内部知识资产（文档、常见问题解答、维基），构建高效的检索体验，并使用检索增强生成（RAG）技术改进模型响应。
- 研究、测试和改进高质量提示，以提高不同场景下的任务执行、可控性和输出质量。
- 开发概念验证和生产就绪的人工智能功能，例如智能问答、文档生成、摘要和代码辅助——重点关注可用性、可靠性和可衡量的价值。
- 密切关注大型语言模型和多模态人工智能的最新进展，并帮助评估新兴工具和方法，以便在公司内部负责任地采用。
必备要求
- 目前正在攻读计算机科学、数据科学、软件工程、信息系统或相关领域的学位（或通过项目获得同等实践经验）。
- 熟悉 Python 和/或其他用于原型设计和集成的编程语言。
- 对大型语言模型和常见的 GenAI 概念有基本的了解（例如，嵌入、提示设计、评估、幻觉风险）。
- 具备良好的英语沟通能力（书面和口头），并能与跨职能部门的利益相关者进行协作。
- 问题解决、文档编写和测试采用结构化、注重细节的方法。
优先条件
- 具有使用 GenAI 框架或工具包（例如 LangChain、LlamaIndex）进行构建和实施 RAG 管道的经验。
- 熟悉 API 集成、微服务或工作流自动化；熟悉容器（Docker）者优先。
- 了解搜索/检索基础知识（向量数据库、索引、排名）和实际评估方法。
- 具备开发内部工具（聊天机器人、知识助手）的经验，注重安全性、隐私性和访问控制。
- 对多模态人工智能（文本+图像/文档处理）以及优化延迟和成本的技术很感兴趣。
Responsibilities
- Support the deployment, optimization, and improvement of large language and multimodal models in internal "smart" use cases such as knowledge search, content writing, and production assistants.
- Package and integrate large model APIs into internal applications and business workflows to enable intelligent upgrades across functions.
- Help organize and structure internal knowledge assets (documents, FAQs, wikis), build efficient retrieval experiences, and improve model responses using retrieval-augmented generation (RAG) techniques.
- Research, test, and refine high-quality prompts to improve task execution, controllability, and output quality across different scenarios.
- Develop proof-of-concept and production-ready AI features such as intelligent Q&A, document generation, summarization, and code assistance—with a focus on usability, reliability, and measurable value.
- Stay up to date with the latest advances in large language models and multimodal AI, and help evaluate emerging tools and methods for responsible adoption within the company.
Must Have
- Currently pursuing a degree in computer science, data science, software engineering, information systems, or a related field (or equivalent practical experience through projects).
- Familiarity with Python and/or other programming languages for prototyping and integration.
- Basic understanding of large language models and common GenAI concepts (e.g., embeddings, prompt design, evaluation, hallucination risks).
- Good English communication skills (written and verbal) and ability to collaborate with cross-functional stakeholders.
- Structured, detail-oriented approach to problem-solving, documentation, and testing.
Preferred
- Experience building and implementing RAG pipelines using GenAI frameworks or toolkits (e.g., LangChain, LlamaIndex).
- Familiarity with API integration, microservices, or workflow automation; experience with containers (Docker) is a plus.
- Knowledge of search/retrieval fundamentals (vector databases, indexing, ranking) and practical evaluation methods.
- Experience developing internal tools (chatbots, knowledge assistants) with a focus on security, privacy, and access control.
- Interest in multimodal AI (text + image/document processing) and techniques for optimizing latency and cost.
公司简介
艾默生（Emerson）是一家全球性的自动化技术与软件公司，成立于 1890 年，总部位于美国密苏里州圣路易斯。业务涵盖流程自动化（控制系统、阀门、测量与分析仪表）、离散自动化，以及收购 NI（美国国家仪器）后形成的测试与测量业务，服务能源、化工、电力、生命科学、半导体等行业。