Rewrite
## About the role
- Design and build automated pipelines for converting reality capture data (photogrammetry, LiDAR point clouds, depth camera scans, 360-degree imagery) into physics-ready 3D simulation assets with accurate geometry, collision meshes, material properties, and articulation definitions.
- Develop and maintain toolchains for physics-ready 3D asset generation, including mesh optimization, UV unwrapping, PBR material assignment, collision hull generation, mass/inertia parameter estimation, and annotation of semantic and functional properties.
- Integrate reality capture hardware and software workflows (e.g., NeRF, Gaussian splatting, structured light scanning, photogrammetry reconstruction) with the simulation platform's asset ingestion pipeline.
- Build 3D reconstruction workflows that enable rapid creation of simulation environments from real-world facility scans, supporting robotics deployment planning, testing, and validation.
- Create and maintain asset toolchains supporting industry-standard formats (USD/OpenUSD, glTF, FBX, OBJ) with appropriate physics and simulation metadata for import into robotics simulation engines.
- Develop synthetic data generation pipelines that leverage high-fidelity 3D assets to produce training data for perception, manipulation, and navigation models, including domain-randomized variations of materials, lighting, object placement, and camera viewpoints.
- Collaborate with robotics engineers, ML researchers, and perception scientists to define asset fidelity requirements, validate simulation-to-reality visual and physical accuracy, and iterate on asset quality based on downstream model performance.
- Implement quality assurance and validation workflows for 3D assets, including automated checks for mesh integrity, physics parameter consistency, rendering fidelity, and simulation stability.
- Review code and technical designs to ensure adherence to team standards for 3D pipeline performance, asset management, and data integrity.
- Remain current in 3D reconstruction, neural rendering, and asset generation research, proactively evaluating new techniques (e.g., generative 3D models, neural radiance fields, 3D Gaussian splatting) for integration into the platform.
## Requirements
- Bachelor's Degree in Computer Science or related technical field AND 4+ years technical engineering experience with coding in languages including, but not limited to, C, C++, C#, Java, JavaScript, or Python
- Master's Degree in Computer Science or related technical field AND 6+ years technical engineering experience with coding in languages including, but not limited to, C, C++, C#, Java, JavaScript, or Python
- OR Bachelor's Degree in Computer Science or related technical field AND 8+ years technical engineering experience with coding in languages including, but not limited to, C, C++, C#, Java, JavaScript, or Python
- OR equivalent experience.
- Experience with 3D reconstruction pipelines, photogrammetry, LiDAR processing, or neural scene representation methods (NeRF, 3D Gaussian splatting).
- Proficiency with 3D content creation and processing tools (Blender, Fusion, Maya, Houdini, MeshLab, Open3D, or equivalent).
- Experience with USD/OpenUSD, OBJ, STL, or equivalent 3D interchange formats and their integration with simulation or rendering engines.
- Background in physically based rendering (PBR), material authoring, or real-time graphics pipeline development.
- Experience building synthetic data generation pipelines for training computer vision or perception models.
- Familiarity with robotics simulation platforms (Isaac Sim, Gazebo, MuJoCo, or equivalent) and their asset requirements.
- Understanding of physics parameter estimation, collision geometry optimization, and articulated object modeling for simulation fidelity.
Sourced via microsoft · Listed on CareerPlan, which tracks 70,000+ jobs from 20+ sources.
