Autonomy Software Engineering Internship, World Modeling

Role Overview

Location

• Houston, Texas, onsite.

Your Role

• Design and implement autonomy and behavior-coordination algorithms for an industrial humanoid platform.
• Integrate robot skills into autonomous behaviors with manipulation, locomotion, and perception engineers.
• Implement data-curation pipelines for real-world industrial tasks.
• Develop synthetic-data generation pipelines for humanoid world-model post-training.
• Build autonomous-behavior evaluation, benchmarking, and regression-testing pipelines.
• Turn industrial video footage into structured, high-quality training data.
• Design reasoning benchmarks for grounding language queries in industrial scenes.

We're Looking For

• Be pursuing or recently graduated from a relevant Robotics, Computer Science, Machine Learning, Electrical Engineering, or related degree program.
• Have hands-on PyTorch deep-learning model training, fine-tuning, or evaluation experience.
• Have strong Python skills for processing scripts, training loops, and evaluation harnesses.
• Know at least one vision-language, large language, or video-understanding model.
• Have experience with data ingestion, filtering, labeling, deduplication, and dataset splitting.
• Be comfortable with Linux, Git, Docker, and cloud GPU infrastructure.
• Be self-directed on open-ended research problems.

Preferred Qualifications

• Experience with NVIDIA Cosmos or comparable world foundation models.
• Exposure to NVIDIA Isaac Sim, Omniverse, or OpenUSD.
• Familiarity with humanoid or mobile-manipulator robotics and ROS or ROS 2.
• Experience with motion capture, egocentric video, or first-person dataset construction.
• Exposure to behavior trees, task and motion planning, or LLM-driven planning.
• Familiarity with NVIDIA Isaac Lab, NIM microservices, or Hugging Face training.
• Prior physical AI, embodied reasoning, or robot-learning internship, research, or open-source work.

Internship Cohorts

• Fall 2026 cohort runs from August through December.
• Spring 2027 cohort runs from January through May.
• Summer 2027 cohort runs from May through August.