Robotics Real-to-Sim 3D Reconstruction Intern

Field AI • Irvine, CA

14d • Hybrid

About The Position

Field AI is transforming how robots interact with the real world. We are building risk-aware, reliable, and field-ready AI systems that address the most complex challenges in robotics, unlocking the full potential of embodied intelligence. We go beyond typical data-driven approaches or pure transformer-based architectures, and are charting a new course, with already-globally-deployed solutions delivering real-world results and rapidly improving models through real-field applications. Field AI — Pioneering Reliable Autonomy in the Real World We are seeking a Robotics Real-to-Sim Reconstruction Intern who is passionate about 3D scene understanding, simulation technologies, and developing workflows that integrate real sensor data into rich virtual environments. If you enjoy working with modern vision tools, high-fidelity simulation engines, and cutting-edge neural reconstruction methods — while getting hands-on experience with real robotics data — this internship is a chance to contribute directly to core autonomy pipelines.

Requirements

• Strong interest in robotics, computer vision, 3D reconstruction, and simulation.
• Experience with camera data (mono and stereo), LiDAR data, or multimodal perception .
• Familiarity with Gaussian Splatting, NeRFs, or related neural reconstruction approaches .
• Working knowledge of Isaac Sim or other simulation frameworks .
• Comfortable programming in Python (C++ and ROS/ROS2 experience is a plus, not required).
• Ability to run experiments, work with datasets, and iterate quickly in an application-focused project.

Nice To Haves

• Experience with vision transformers, depth estimation, or domain adaptation .
• Familiarity with robot odometry or SLAM outputs for guiding reconstructions.
• Background in robotics field testing or sensor calibration .
• Exposure to GPU acceleration, PyTorch, or custom neural pipelines .

Responsibilities

• Work with real robot camera and LiDAR datasets to test end-to-end reconstruction pipelines.
• Apply techniques such as Gaussian Splatting, NeRF-style reconstruction, and vision transformer–based models to create realistic scene representations.
• Develop real-to-sim workflows that convert raw sensor data into USD environments for simulation platforms, such as Isaac Sim .
• Experiment with NVIDIA tools and modern GPU workflows to validate reconstruction quality and simulation utility.
• Collaborate with autonomy and hardware teams to align reconstructed environments with real robot behavior and field constraints.

What This Job Offers

• Career Level: Intern
• Education Level: No Education Listed
• Number of Employees: 11-50 employees