NVIDIA's Jaiveer Singh discusses how Isaac ROS, an open source robotics software platform built on CUDA-accelerated libr

Jaiveer Singh, a robotics software engineer leading NVIDIA's Isaac ROS team, emphasizes that building practical robots requires focus on infrastructure before spectacle. Isaac ROS, built on the open source ROS 2 framework, brings CUDA-accelerated libraries and AI models to developers creating autonomous mobile robots, manipulation systems and humanoids. Singh's goal is to ensure everyone feels included in the robotics future.

Singh's robotics journey began in middle school with LEGO Mindstorms kits. After excelling in robotics competitions through high school, he studied electrical engineering, computer science and business at UC Berkeley, then joined NVIDIA full time following an internship with the robotics team. Notably, the Isaac ROS work he now leads originated as his intern project. When the team proposed releasing robotics software as open source using the NVIDIA Jetson platform and CUDA libraries, Singh recalled asking whether there would be value. "The answer was, of course, yes," he said, "because developers always want to be able to unlock the full power of their GPUs."

The robotics field has long balanced extraordinary imagination with physics-bound realities. While viral robot videos spread quickly, building systems that work reliably across different sensors, platforms, factories and labs requires sustained effort. Isaac ROS provides a complete stack including simulation, training, accelerated computing, AI models, middleware and edge deployment. The platform supports manipulation, mobility and humanoids, offering packages for perception, object detection, mapping, collision detection and motion planning. It runs on workstations, NVIDIA DGX Spark personal AI supercomputers and NVIDIA Jetson edge systems.

"Compared with the original Isaac SDK, Isaac ROS is completely modular," Singh said. "We ship the software like a bunch of LEGO bricks — you get to assemble them however you want, and you can easily combine our packages with existing ROS code written by you or others in the global robotics community."

Open source is valuable, Singh explained, because it gives developers confidence to build on a platform they can inspect, adapt and trust. As the robotics landscape shifts rapidly, developers need assurance that the platform will remain available to modify and improve years into the future. This confidence matters especially as humanoid robots have moved from science fiction to an active engineering frontier. Isaac ROS has been enhanced for developers using AI agents and for humanoid systems requiring end-to-end software stacks.

NVIDIA's established presence in robotics initially attracted Singh to the company. "NVIDIA was here and working on this problem before anybody else thought it was important," he said. "We already had a stake in the ground."

Singh views open source as a way of sharing both confidence and responsibility. When a robotics startup builds on a closed system, it must trust the system will meet its future needs. With open software, developers can inspect code, change it, contribute fixes and carry it forward. One company's bug fix becomes another company's acceleration. "When more people can build robots," Singh concluded, "the future gets here faster."