Tech Stack
AWSAzureCloudGoogle Cloud PlatformPythonPyTorchTensorflow
About the role
- Design & Implement CV Models – Build computer vision systems for object detection, damage assessment, image segmentation, and measurement extraction from aerial, drone, and ground-level imagery
- Advance 3D Reconstruction – Develop algorithms to generate accurate 3D models of homes and property structures using images, LiDAR, and multimodal data sources
- Innovate in Real-World Applications – Create real-time or near-real-time solutions for mobile and web platforms to power field inspections
- Leverage Multimodal Data – Integrate data from RGB imagery, aerial/satellite sources, and LiDAR to produce richer, more reliable insights
- Collaborate & Influence – Partner with product, engineering, and insurance experts to translate technical outputs into actionable business decisions
- Drive Continuous Improvement – Stay at the forefront of computer vision and machine learning advancements, and apply emerging techniques to enhance accuracy, scalability, and efficiency
Requirements
- 5+ years in computer vision, machine learning, or applied AI roles, preferably in real-world inspection, insurance, construction, or geospatial domains
- Bachelor’s, Master’s, or PhD degree in Computer Science, Electrical Engineering, Mathematics, Physics, or related technical field
- Deep Learning for 2D image classification, object recognition and semantic/instance segmentation
- Strong Background in multimodal data fusion (e.g., RGB, aerial imagery, satellite, LiDAR)
- Proficiency in Python, TensorFlow/PyTorch, OpenCV and strong applied math skills in geometry, linear algebra and optimization
- Strategic Mindset-Ability to see beyond the algorithm, anticipating business needs and shaping product direction through AI-driven insights
- Nice to Have: Experience with insurance, property tech or construction tech solutions
- Nice to Have: Familiarity with cloud-based ML deployment pipelines (AWS/GCP/Azure)
- Nice to Have: Experience building interactive visualization tools for inspection data or 3D reconstructions