AI Engineer (Point Cloud → CAD) | Python, Computer Vision, Open3D | Part-Time

Overview We're looking for an experienced AI Engineer to help build an automated pipeline that converts 3D point cloud data into CAD drawings. Our surveying team captures millions of 3D laser scan points every week. Today, much of the drafting process is manual. We want to automate this by using AI and computer vision to detect real-world structures (walls, doors, windows, edges, etc.) and generate clean vector-based CAD output. This is a hands-on engineering role focused on solving real-world geometric and computer vision problems. Responsibilities Build AI and computer vision models for processing 3D point cloud data Develop pipelines that convert E57/LAS/LAZ files into structured CAD linework Detect architectural features such as walls, doors, windows, floors, ceilings, and edges Generate clean vector output (DWG/DXF or equivalent) Improve accuracy through validation, testing, and quality assurance Collaborate with our engineering team to iterate on algorithms and system architecture Document technical decisions and implementation details Required Skills Strong Python development experience Hands-on experience working with point cloud data (E57, LAS/LAZ, PLY, PCD, or similar) Experience with computer vision and 3D geometry processing Experience using libraries such as Open3D, PCL, or similar Experience building machine learning or deep learning models for spatial or geometric data Understanding of CAD formats (DWG, DXF, or similar) Ability to write clean, maintainable production-quality code Good communication skills in English Nice to Have Experience with CAD automation AWS, Azure, or GCP experience Knowledge of Scan-to-BIM or automated drafting workflows Experience with segmentation or object detection in point clouds Experience with PyTorch or TensorFlow Familiarity with AI-assisted development tools such as Claude Code or GitHub Copilot Project Details Part-time contract Approximately 4–5 hours per day Must be available to work during Philippine Time (GMT+8) for collaboration Long-term opportunity for the right candidate