AI Infrastructure and Frameworks Intern, Cosmos Lab - 2027
Core
Develop training and post-training systems for advanced Physical AI models, including world foundation models and robot policies, connecting training, inference, and evaluation with simulation and real-world robot interaction.
Role type
Intern, AI Infrastructure Engineer
Builds
Training infrastructure, RL infrastructure, simulation/robotics integration pipelines
Domain
AI Infrastructure, Physical AI, Robotics
Deliverable
production ML models
Required skills
Python, distributed execution, memory management, data movement, systems fundamentals
Preferred skills
GPU profiling, C++/CUDA development, open-source contributions, research in ML systems
Technologies
NVIDIA GPU infrastructure, simulation environments, robot interfaces
Responsibilities
Develop and optimize training infrastructure for advanced Physical AI world models; Build Physical AI post-training and RL infrastructure; Improve efficiency and scalability across training, inference, simulation, and evaluation; Analyze and optimize system performance using profiling and benchmarking.
