数据挖掘工程师（汽车行业）

该职位来源于猎聘 python+VLM+汽车行业+图像类经验（车辆数据集）

我们正在招聘一名数据挖掘工程师，负责面向汽车场景开发数据驱动解决方案，重点围绕大规模车辆传感器数据进行挖掘、分析和应用。理想候选人应具备汽车行业实践经验，熟悉摄像头、激光雷达、毫米波雷达等车辆传感器，并具备开源视觉语言模型（VLM）部署经验，以及使用 Qwen、Doubao 等云端 VLM API 的经验。 Key Responsibilities | 主要职责

• Mine, analyze, and structure large-scale automotive datasets from vehicle sensors, test fleets, simulation systems, and production data sources.
• Develop data pipelines for collection, cleaning, deduplication, labeling, quality assessment, and feature extraction.
• Analyze camera, lidar, radar, and multi-sensor data to identify driving scenarios, corner cases, perception failures, and data coverage gaps.
• Build data mining workflows that support autonomous driving, ADAS, perception model development, evaluation, and validation.
• Deploy and evaluate open-source Vision-Language Models for automotive data understanding, scenario tagging, captioning, retrieval, and quality inspection.
• Integrate cloud-based VLM APIs such as Qwen and Doubao into scalable data processing and analysis workflows.
• Collaborate with perception, data engineering, product, and vehicle testing teams to translate business and engineering needs into data mining solutions.
• 挖掘、分析并结构化来自车辆传感器、测试车队、仿真系统及量产数据源的大规模汽车数据集。
• 开发数据采集、清洗、去重、标注、质量评估及特征提取流程。
• 分析摄像头、激光雷达、毫米波雷达及多传感器数据，识别驾驶场景、长尾案例、感知失败问题及数据覆盖缺口。
• 构建支持自动驾驶、ADAS、感知模型开发、评估与验证的数据挖掘流程。
• 部署并评估开源视觉语言模型，用于汽车数据理解、场景标注、图文描述、检索及质量检查。
• 将 Qwen、Doubao 等云端 VLM API 集成到可扩展的数据处理和分析流程中。
• 与感知算法、数据工程、产品及车辆测试团队协作，将业务和工程需求转化为数据挖掘解决方案。 Required Qualifications | 任职要求
• Hands-on experience in the automotive industry, preferably in autonomous driving, ADAS, intelligent cockpit, connected vehicles, or vehicle data platforms.
• Practical experience working with vehicle sensor data, including cameras and lidar or radar.
• Experience processing, mining, and managing large-scale datasets, including structured, unstructured, image, video, and sensor data.
• Experience deploying open-source Vision-Language Models for real-world data analysis or production workflows.
• Experience using cloud-based VLM APIs such as Qwen, Doubao, or similar multimodal model services.
• Strong Python programming skills and familiarity with data processing frameworks such as Pandas, Spark, PyTorch, or similar tools.
• Ability to design scalable data mining pipelines and evaluate data quality, relevance, coverage, and usability.
• Good communication skills and ability to collaborate with cross-functional engineering teams.
• 具备汽车行业实际经验，优先考虑自动驾驶、ADAS、智能座舱、车联网或车辆数据平台相关经验。
• 具备车辆传感器数据处理经验，包括摄像头以及激光雷达或毫米波雷达数据。
• 具备大规模数据集处理、挖掘和管理经验，包括结构化、非结构化、图像、视频及传感器数据。
• 具备开源视觉语言模型在真实数据分析或生产流程中的部署经验。
• 具备使用 Qwen、Doubao 或同类云端多模态模型服务的经验。
• 具备扎实的 Python 编程能力，熟悉 Pandas、Spark、PyTorch 或类似数据处理框架。
• 能够设计可扩展的数据挖掘流程，并评估数据质量、相关性、覆盖度及可用性。
• 具备良好的沟通能力，能够与跨职能工程团队协作。 Preferred Qualifications | 加分项
• Experience with autonomous driving perception datasets, scenario mining, corner-case discovery, or data closed-loop systems.
• Familiarity with multimodal retrieval, image/video understanding, data tagging, active learning, or weak supervision.
• Experience with distributed data processing, cloud storage, GPU-based model serving, Docker, Kubernetes, or MLOps workflows.
• Knowledge of automotive data formats, calibration, synchronization, sensor fusion, or perception evaluation metrics.
• Contributions to open-source VLM, multimodal AI, data mining, or autonomous driving projects.
• 具备自动驾驶感知数据集、场景挖掘、长尾案例发现或数据闭环系统经验。
• 熟悉多模态检索、图像/视频理解、数据标注、主动学习或弱监督方法。
• 具备分布式数据处理、云存储、基于 GPU 的模型服务、Docker、Kubernetes 或 MLOps 流程经验。
• 了解汽车数据格式、标定、时间同步、传感器融合或感知评估指标。
• 有开源视觉语言模型、多模态 AI、数据挖掘或自动驾驶项目贡献经验