Machine Learning Engineer

免费注册后解锁"你的匹配分 vs 高分画像"与个人技能缺口。

Vantor is forging the new frontier of spatial intelligence, helping decision makers and operators navigate what’s happening now and shape what’s coming next. Vantor is a place for problem solvers, changemakers, and go-getters—where people are working together to help our customers see the world differently, and in doing so, be seen differently. Come be part of a mission, not just a job, where you can: Shape your own future, build the next big thing, and change the world. To be eligible for this position, you must be a U.S. Person, defined as a U.S. citizen, permanent resident, Asylee, or Refugee. Export Control/ITAR: Certain roles may be subject to U.S. export control laws, requiring U.S. person status as defined by 8 U.S.C. 1324b(a)(3). Please review the job details below. Role Overview Vantor is seeking a Machine Learning Engineer to build the systems that turn raw satellite imagery into trusted, well-labeled data, and to develop machine learning algorithms to turn that data into models. Your work sets the quality, speed, and traceability of the datasets behind Vantor 's geospatial and machine learning products. You will work closely with internal domain experts, external annotation partners, and Vantor 's Machine Learning and Software Engineering teams to design robust, scalable, low-latency solutions. On the platform side, you will build pipelines for large geospatial imagery and metadata, including sensor, acquisition, and geolocation attributes, and move data reliably between annotation, storage, and model training systems. You will also track data lineage and versioning so every label can be traced back to its source image, guideline version, and annotator. On the machine learning side, you will develop, adapt, and deploy models that generate proposed annotations, cutting manual labeling effort and turnaround time. You will build and operate production inference pipelines, evaluate model performance against labeled data, and use model feedback to decide which imagery should be labeled next. This is a hands-on role for an engineer who is as comfortable shipping a reliable data service as training and deploying a computer vision model, and who can connect the two into a single, continuously improving system. Minimum Requirements 5+ years of relevant experience in machine learning engineering, data engineering, backend software engineering, or a closely related technical role. Hands-on experience training, fine-tuning, or adapting deep learning models using a modern ML framework. Experience deploying machine learning models into production, including building batch or real-time inference pipelines and managing model versions. Hands-on experience building and operating services on a major cloud platform, including managed compute, databases, storage, and identity and access management. Experience building and operating workflow orchestration for data or ML pipelines. Experience owning containerized services, CI/CD pipelines, and infrastructure-as-code. Proven ability to support production systems, including debugging, observability, secure configuration, and incident response. Strong Python application development skills. Bachelor's degree in Computer Science, Machine Learning, Data Engineering, Geospatial Science or GIS, Remote Sensing, or a related discipline, or equivalent demonstrated experience. Preferred Qualifications Experience with Google Cloud Platform services, including Cloud Run, Cloud SQL, Cloud Storage, IAM, and service account management. Experience with Vertex AI, including Vertex AI Pipelines, training jobs, and model deployment, or with similar tools such as Kubeflow Pipelines. Experience with PostgreSQL and PostGIS, including spatial data modeling and managing schema migrations in production. Experience working with satellite or aerial imagery, including raster formats such as GeoTIFF, coordinate reference systems, and tools such as GDAL, Rasterio, Shapely, Fiona, and GeoPandas. Experience developing com

Vantor正在开拓空间智能的新前沿，帮助决策者和运营人员洞察当下正在发生的事，并塑造即将到来的未来。Vantor是解决问题者、变革者和行动者的家园——人们在这里携手合作，帮助客户以不同的方式看世界，并因此被以不同的方式看待。来加入一项使命，而不仅仅是一份工作，在这里你可以：塑造自己的未来，打造下一个重大事物，并改变世界。要获得此职位的资格，您必须是美国人士，定义为美国公民、永久居民、庇护者或难民。出口管制/ITAR：某些职位可能受美国出口管制法律约束，要求具备8 U.S.C. 1324b(a)(3)所定义的美国人士身份。请查看下方的职位详情。职位概述Vantor正在寻找一名机器学习工程师，负责构建将原始卫星影像转化为可信、标注良好的数据的系统，并开发机器学习算法将这些数据转化为模型。您的工作将决定Vantor地理空间和机器学习产品背后数据集的质量、速度和可追溯性。您将与内部领域专家、外部标注合作伙伴以及Vantor的机器学习和软件工程团队紧密合作，设计稳健、可扩展、低延迟的解决方案。在平台方面，您将为大规模地理空间影像和元数据构建管道，包括传感器、采集和地理定位属性，并在标注、存储和模型训练系统之间可靠地传输数据。您还将跟踪数据血缘和版本管理，使每个标签都能追溯到其源影像、指南版本和标注者。在机器学习方面，您将开发、调整和部署生成建议标注的模型，减少人工标注工作量和周转时间。您将构建和运营生产推理管道，根据标注数据评估模型性能，并利用模型反馈来决定接下来应标注哪些影像。这是一个动手实践的角色，适合一位既能交付可靠的数据服务，又能训练和部署计算机视觉模型的工程师，并且能够将两者连接成一个持续改进的单一系统。最低要求 5年以上机器学习工程、数据工程、后端软件工程或密切相关的技术角色的相关经验。具有使用现代ML框架训练、微调或调整深度学习模型的实践经验。具有将机器学习模型部署到生产环境的经验，包括构建批处理或实时推理管道以及管理模型版本。具有在主流云平台上构建和运营服务的实践经验，包括托管计算、数据库、存储以及身份和访问管理。具有为数据或ML管道构建和运营工作流编排的经验。具有负责容器化服务、CI/CD管道和基础设施即代码的经验。具有支持生产系统的 proven 能力，包括调试、可观测性、安全配置和事件响应。强大的Python应用开发技能。计算机科学、机器学习、数据工程、地理空间科学或GIS、遥感或相关学科的学士学位，或同等证明的经验。优先资格 具有Google Cloud Platform服务经验，包括Cloud Run、Cloud SQL、Cloud Storage、IAM和服务账户管理。具有Vertex AI经验，包括Vertex AI Pipelines、训练作业和模型部署，或具有类似工具如Kubeflow Pipelines的经验。具有PostgreSQL和PostGIS经验，包括空间数据建模和生产环境中管理模式迁移。具有处理卫星或航空影像的经验，包括GeoTIFF等栅格格式、坐标参考系统以及GDAL、Rasterio、Shapely、Fiona和GeoPandas等工具。具有开发com的经验

⚑ 岗位信息收录自公开招聘页（Himalayas），以雇主官网为准