NVIDIA Learning Pathway

Isaac Sim basics, building robots in simulation, sensor configuration, and ROS 2 integration. The foundational course for understanding the simulation platform.

USD fundamentals, scene composition and layering, Python scripting for USD, and asset pipeline workflows. A multi-part learning path covering the 3D format standard.

Omniverse platform architecture, Kit SDK, building extensions, and working with the Omniverse ecosystem. Covers the full development platform.

GPU programming basics, writing CUDA kernels, memory management and optimization, and accelerated computing paradigms with Python.

A multi-course path covering robot perception, manipulation, and autonomous navigation. Provides broader robotics context beyond simulation.

Digital twin concepts, USD-based workflows, simulation pipelines, and creating high-fidelity virtual replicas of physical environments.

Reinforcement learning environments, task design and reward shaping, policy training pipelines, and sim-to-real transfer methods.

Sparse voxel data structures on GPU, efficient spatial queries, NanoVDB integration, and large-scale volumetric data management.

Retrieval-augmented generation, LLM agent design, prompt engineering, and building AI-powered assistants for domain-specific tasks.

ROS 2 Humble fundamentals, nodes and topics, robot communication, and hardware driver integration for real-world deployment.

Neural network architectures, training and optimization, inference pipelines, and practical deep learning applications.

Acoustic ray tracing, sound propagation modeling, transmission loss calculations, and environmental input file formats.