What You Should Know About Content Aggregation | OpenUSD Community Office Hour
Learn practical OpenUSD content aggregation patterns, from encapsulation and default prims to reference vs payload, variant workflows, lofting, and model kinds. Part of a 6-week series prepping for the OpenUSD Developer Certification (free at GTC). You will learn how to structure assets and scenes for: - Legibility, modularity, performance, and navigability (the SD4 principles) - Clean interfaces using default prims and encapsulation conventions - Faster iteration with the reference + payload pattern - Parameterization with variant sets and primvars - Parallel collaboration with workstreams (layering strategy) - Lofting (exposing important metadata without loading heavy payloads) - Model hierarchy using model kinds (component, assembly, group) 🎟️ Reserve your #NVIDIAGTC 2026 certification spot: https://nvda.ws/4riFpr0 📅 Add the OpenUSD prep course to your calendar: https://nvda.ws/4kEbQhi 🎥 Watch the full prep series: https://youtube.com/playlist?list=PL3jK4xNnlCVf3HuZD4qOWlKlouJyh6Prb&si=JXpXQbFD7-snbq84 ---------------------------------------------------------------------------- Key Moments: 0:03 Intro and welcome 0:20 Topic overview: Content aggregation in OpenUSD 0:59 Guests: Haley and Maddie intros 2:19 OpenUSD Developer Certification and Learn OpenUSD 3:32 Series framing: 6-week exam prep and recordings 5:00 Haley background and goals for the session 7:22 Agenda overview and common workflow problems 8:06 SD4 principles: legibility, modularity, performance, navigability 9:03 Quiz 1: Naming and public vs internal clarity 11:50 Moving assets safely: paths and encapsulation 15:06 Default prims and asset interfaces 15:52 Reference vs payload: avoid loading heavy geometry 16:29 Pixar Kitchen Set demo context 17:39 Reference and payload boundary walkthrough 20:43 Quiz 2: What belongs in reference vs payload 22:29 Asset parameterization with variants (Houdini Solaris) 24:32 Tooling example: variant comparison preview 26:00 Primvars as parameterization (interpolated values) 27:04 Workstreams: layering for parallel collaboration 27:17 Quiz 3: Best structure for multi-artist assets 30:11 Lofting: exposing data from payload into reference 31:48 Model hierarchy and model kinds overview 32:14 Kinds explained: component vs assembly vs group 33:38 Pixar Kitchen hierarchy example with kinds 35:46 Q&A: Should group/assembly be separate files? 40:27 Q&A: Why define kinds? 41:30 Quiz 4: Model hierarchy contract violation 42:58 Wrap up and thanks 43:24 Community resources: LinkedIn, study group, Discord 44:11 Q&A: Do you need to be a programmer? 45:22 Q&A: Relative path authoring behavior 46:30 Q&A: Houdini to Omniverse interoperability discussion 50:51 Closing remarks from speakers 51:49 GTC 2026 overview and certification details 52:09 Physical AI Day talks at GTC 53:25 Hands-on trainings and labs at GTC 54:33 Get OpenUSD certified at GTC (free) 55:29 Physical AI developer meetup 56:33 Next week: Customizing USD ---------------------------------------------------------------------------- 📆 Check out the full calendar for all of our upcoming events → https://nvda.ws/3JqaWnA ---------------------------------------------------------------------------- ⬇️Get Started → https://nvda.ws/4cZAZO1 👀Explore OpenUSD → https://nvda.ws/3CeozBQ 👥Join the Community → https://nvda.ws/3ZMfc6e #OpenUSD #USD #3DPipeline #ContentAggregation #SceneGraph #DigitalContentCreation #VariantSets #USDPayloads #USDReferences #AssetPipeline #Simulation #Robotics #GTC2026