> Markdown version of [/jobs/ext/3811349-senior-ml-engineer-computer-vision](https://www.wearedevelopers.com/jobs/ext/3811349-senior-ml-engineer-computer-vision). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # (Senior) ML Engineer - Computer Vision - **Company:** credium GmbH - **Location:** Augsburg, Germany - **Experience:** Expert - **Contract:** Permanent contract - **Skills:** 3D Scanning, Geographic Information Systems, Artificial Intelligence, Computer Vision, Microsoft Azure, Batch Processing, Cyber Security, Continuous Integration, Data Auditing, Data Validation, Information Engineering, Python (Programming Language), PostGIS, Quantum GIS (QGIS), Azure Machine Learning, Management of Software Versions, Data Processing, Agentic-AI, Gaussian, Information Technology, Machine Learning Operations, Drift Detection, Lidar, Raster Graphics - **Published:** October 18, 2026 - **Apply:** https://www.careerjet.de/jobad/de803966ea18a5502868efa455d78a8667 ## About the Role You aren't just looking for a job; you're looking for a stage for your entrepreneurial mindset and your drive to get things moving-and you want to be part of an ambitious team aiming to achieve great things together? Then we should talk., * You hold a degree or equal experience in Computer Science, Data Science/Engineering, Geoinformatics, Remote Sensing or a related field, or comparable working experience * You have several years of professional experience bringing ML systems into production (MLOps, CI/CD, monitoring), ideally on Azure * You have hands-on experience with computer vision in production - you can evaluate, fine-tune and operate models confidently * You have working knowledge of 3D vision data: processing point clouds (e.g. LiDAR, LAS/LAZ), meshes, Gaussian Splats, Voxels, and 3D geometries with tools such as PDAL, Open3D, trimesh or PyVista - and ideally first experience with 3D data formats like CityGML or 3D Tiles * You have experience with geospatial data and aerial imagery (PostGIS, QGIS, raster/vector processing) or with photogrammetry and 3D reconstruction and know what georeferencing is * You possess strong Python and solid data engineering skills (pipelines, batch processing, online inference, data quality checks, CI/CD) * You have practical experience with modern AI tooling: foundation models, auto-labeling, vision platforms (e.g. Roboflow) - experience with agentic AI workflows is a strong plus * You think pragmatically and economically: you enjoy weighing build-vs-buy options and are happy to choose open models, platforms or existing data when that is the smarter path for the product * You have experience with labeling operations and data quality assurance, or the structured mindset to set them up * You communicate clearly, document well and enjoy sharing your knowledge and enabling others (working students, labeling team, colleagues) * You are highly motivated, hands-on and comfortable with the pace, ambiguity and ownership of a startup environment * (Bonus) Knowledge of IT security principles in the context of data handling ## Description We enable mortgage banks, property owners, and real estate professionals to understand buildings better and faster - from floor plans to energy efficiency, from renovation potential to solar readiness. We're a VC-funded startup with major banks as customers and a purpose that matters. Aufgaben Why this role matters What you'll do * Own the lifecycle of our productive vision models end-to-end: training playbooks, evaluation with gold sets and per-class metrics, versioning, release gates and drift monitoring * Harden and operate our batch inference and geospatial postprocessing pipelines on Azure ML, treating every inference run as a controlled, cost-aware data product release * Work with 3D vision data - point clouds (LiDAR), meshes and 3D building geometries - and help us combine 2D imagery with 3D data into richer building intelligence * Steer our labeling operations together with our experienced external labeling team: you define specifications, QA sampling and acceptance criteria and empower the team to deliver great data quality * Shape our broader AI direction, including the use of agentic AI and automation to keep our processes lean and our team focused * Collaborate closely with our data engineers, data scientists and product manager, sharing knowledge openly so vision expertise grows across the whole team This role is perfect for someone who thrives on responsibility, loves building clear processes and playbooks, and takes pride in making AI reliable, sustainable and valuable in production. Profil Don't worry if you don't fulfill all criteria. Everyone's got a focus.