🇩🇪 Berlin, Germany · 6h ago
Computer Vision Engineer (m/f/d) | Berlin, Hybrid
Airteam Aerial Intelligence
LinkedInmidEnglish-friendly
Airteam is a Berlin-based SaaS startup with an AI-powered cloud platform for roofers and solar planners. As a mid-level Computer Vision Engineer, you'll build 2D-to-3D object detection and automatic geometry reconstruction from drone data. Your work helps roofers and solar planners make informed decisions and drives a climate-neutral product.TasksDevelop and implement computer vision algorithms for 2D-to-3D object detection using drone dataIntegrate ML models into our platform together with the engineering teamsBuild end-to-end ML pipelines with GCP, Python, Argo, PyTorch and MLflowImplement data preprocessing, model training, deployment and monitoringVisualize and communicate findings to stakeholdersRequirements2-4 years of commercial experience in data science/ML, including Computer Vision and Deep LearningExperience across the full ML lifecycle (preprocessing, training, evaluation, deployment, monitoring), incl. conda and MLflow or comparable toolsSolid software engineering skills: Python, Git/GitHub, Docker/Kubernetes, cloud platformsStrong problem-solving, communication and teamwork skills, plus a willingness to learn new tools quickly**Nice-to-have**Experience in construction or a related industryExperience with our stack: GCP, Vertex AI, PyTorch, Conda, MLflow, GitHub, Argo WorkflowsExperience building and maintaining ML pipelinesA goal-oriented, "just do it" mentalityBenefitsHybrid work, workations, and remote work by contractModern working space in the heart of Berlin30% contribution towards your pension savingsRegular team events and workshopsFlat hierarchies, short communication channels, room for your own ideasJoin Airteam in Berlin to revolutionize drone data with AI! As a Computer Vision Engineer, blend tech innovation with creativity in a dynamic, hybrid work setting. Apply now!Sourced from LinkedIn. Relocantly aggregates public job postings; apply on the original site.