# Develop Lightweight USD Runtimes Faster with AI Agents

Accelerate your physical AI projects using the USD Core Specification.

- **Source:** [Nvidia Developer Blog](https://developer.nvidia.com/blog)
- **Published:** 2026-07-15
- **Topic:** [Data & AI](https://brifing.si/c/data-ai)
- **Original article:** <https://developer.nvidia.com/blog/develop-lightweight-usd-runtimes-faster-with-ai-agents/>
- **On Brifing:** https://brifing.si/s/56bb6677-b0a7-44b9-9225-29addb99548f

## OpenUSD and AI Agents

OpenUSD provides a common scene description language for physical AI. AI agents use the USD Core Specification to create custom runtimes. This approach complements traditional codebase adaptation and enables faster implementation cycles.

## USD Core Specification

The USD Core Specification defines how USD data models are composed and resolved across layer stacks. This specification serves as a precise contract for both humans and AI agents, enabling compliant runtimes.

## nanousd-labs Project

nanousd-labs is published as part of NVIDIA Omniverse Labs. This project allows developers to generate runtimes directly from the USD Core Specification using AI agents, ideal for projects with specific memory and performance requirements.

## Role of AI Agents

AI agents treat the USD Core Specification as a contract. They read the specification section by section, write code to implement each behavior, and validate the output against a test suite derived from the specification, ensuring compliance.

## Mechanical Work and Engineering Decisions

In nanousd-labs, AI agents handle mechanical specification-to-code work, while engineers manage performance and architectural decisions. This division combines the benefits of automation and human expertise.

## Flexible Code Generation

Code generated by AI agents can be regenerated against different constraints such as memory, performance, or language. This flexibility allows developers to create runtimes tailored to their workloads.

## Testing Process

AI agents validate the generated code against a test suite derived from the specification. This process is repeated to ensure the output meets the specification requirements, guaranteeing compliance.

## Developer Control

AI agents operate under developer guidance. Developers control the process of reading the specification section by section, writing code to implement each behavior, and testing it, enabling custom runtime creation.

## Implementation Example

The nanousd-labs project is a real-world example of AI agents using the USD Core Specification to create runtimes. This approach enables fast and compliant implementation cycles in physical AI projects.

## Future Potential

The approach of using AI agents and the USD Core Specification to create runtimes holds significant potential for physical AI projects. It allows developers to quickly develop solutions tailored to specific requirements.

## Sources covering this story

- https://developer.nvidia.com/blog/develop-lightweight-usd-runtimes-faster-with-ai-agents/ — 

---

_Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._