锦鲤求职岗位描述
岗位职责Roles & Responsibilities：
• 数据ETL开发：负责多源异构数据（数据库、日志、API、文件等）的抽取、清洗、转换与加载，构建稳定高效的数据管道（Data Pipeline）
• 数据分析支持：基于业务需求进行探索性数据分析（EDA），产出数据洞察报告，支撑业务决策
• Data Agent 探索与构建：调研并实践基于大语言模型（LLM）的智能数据代理（Data Agent），探索将 ETL 流程、数据质量检测、分析报告生成等环节自动化
• 全生命周期自动化流程建设：参与构建从"需求理解 → 数据采集 → 清洗转换 → 分析建模 → 结果呈现"的端到端自动化开发流程，提升数据工程效率
• 工具链搭建：协助搭建/优化数据开发相关的 Agent 工具（如自动生成 SQL、自动化数据质检、自动报告生成等）
• Data ETL Development: Responsible for the extraction, cleaning, transformation, and loading (ETL) of multi-source heterogeneous data (databases, logs, APIs, files, etc.), building stable and efficient data pipelines.
• Data Analysis Support: Conduct exploratory data analysis (EDA) based on business requirements, producing data insight reports to support business decision-making.
• Data Agent Exploration & Development: Research and implement LLM-based intelligent data agents, exploring the automation of ETL processes, data quality checks, and analytical report generation.
• Full Lifecycle Automation Pipeline Development: Participate in building an end-to-end automated development workflow covering "requirement understanding → data collection → cleaning & transformation → analysis & modeling → results presentation," improving data engineering efficiency.
• Toolchain Development: Assist in building/optimizing Agent tools related to data development (such as automatic SQL generation, automated data quality checks, automated report generation, etc.).
任职要求Qualifications & Skills Required:
Education & major requirement 学历及专业要求
• 计算机科学、软件工程、数据科学、统计学、人工智能等相关专业本科及以上学历
• 熟悉 Python，有 Pandas / NumPy / SQL 实际项目或课程经验 了解基本的数据仓库/数据湖概念（如 ODS/DWD/DWS/ADS 分层）
• 对大语言模型（LLM）、Prompt Engineering、Agent 框架（如 LangChain、LangGraph、AutoGen、CrewAI 等）
• 有学习经历或项目实践者优先 具备较强的逻辑思维能力、自驱力和快速学习能力
• 良好的英文文档阅读能力（大量前沿资料为英文）
• Bachelor's degree or above in Computer Science, Software Engineering, Data Science, Statistics, Artificial Intelligence, or related fields; Class of 2026 graduates.
• Proficient in Python, with hands-on project or coursework experience in Pandas / NumPy / SQL.
• Familiar with basic data warehouse/data lake concepts (e.g., ODS/DWD/DWS/ADS layered architecture).
• Prior learning or project experience with Large Language Models (LLMs), Prompt Engineering, or Agent frameworks (such as LangChain, LangGraph, AutoGen, CrewAI, etc.) is preferred.
• Strong logical thinking ability, self-motivation, and fast learning capability.
• Good English reading comprehension for technical documentation (as much cutting-edge material is published in English).