EmergingMultimodal and generative world models

Generative 3D Researcher

Also called 3D Generative AI Scientist

A Generative 3D Researcher builds models and pipelines that produce spatial assets with controllable structure. They matter because generated geometry only has value if it is valid and works in the renderers and simulators the product depends on.

What they do

They design and train generative models for 3D content, then build the pipeline around them. That means choosing a representation, such as meshes, implicit fields or splats, and deciding how text, images or other inputs steer the output. They write geometry checks for things like watertightness, scale, topology and material consistency. They test outputs in target renderers and simulators, not only in a viewer. They read failures closely, trace them to data, representation or training choices, and run focused experiments to fix them. They also work with engineers who consume the assets downstream.

What a strong hire delivers

In the first months, a strong hire delivers a working generation pipeline from input to asset, with clear controls over structure. They put geometry validation in place so bad outputs are caught automatically. They establish a measure of usability in the target renderers, agree it with the people who use the assets, and show a baseline with a credible plan to improve it. They also leave behind clear notes on what worked and what did not.

How you can tell they are good

Ask them to walk through a 3D generation system they built and where it broke. A strong candidate explains why they chose their representation and what it cost them. They define validity in concrete terms and show how they tested it in a real renderer. They talk about failure cases without being asked. Be wary of people who show only polished renders and cannot say how the assets behave once they are imported elsewhere.

Why it is hard to fill

The work is usually hidden inside other titles, so job titles are a poor guide. Most candidates are strong in one area: multimodal modelling, graphics, or simulation. Few have all three, and fewer still care about validity as much as visual quality. People with real depth are often tied up in research teams, so a search needs to read the work, not the title.

Where to find them

Look at multimodal ML researchers who have moved into 3D, generative modelling researchers working on images or video, and simulation or graphics researchers who have picked up learned models. Some are titled research scientist or research engineer. Others sit in robotics, games, visual effects or digital twin teams. Frontier labs, university groups and applied research startups are all sources.

Indicative pay

Not enough published evidence yet to give a fair range for this role, which is common for work this new. We benchmark it for you as part of a search.

Interview questions for a Generative 3D Researcher

  1. Tell me about a 3D generation system you built. Which representation did you choose, and why?

    A strong answer names the alternatives considered and the trade-offs in quality, control, speed and downstream use. They tie the choice to the target use. Worry if the choice was simply whatever was fashionable or inherited.

  2. How do you define and test geometric validity for generated assets?

    Look for concrete checks such as watertightness, manifold structure, scale, self-intersection and material sanity, run automatically. They should separate validity from visual appeal. Worry if validation is manual or relies on eyeballing renders.

4 more questions, with what a strong answer covers, are in the Role Kit.

Job description you can post

We are building generative systems that produce 3D assets people can actually use. We need a researcher who can take models from idea to a reliable pipeline, with clear control over structure and strong checks on geometry. You will work closely with engineers who use the assets.

What you will do

  • Design and train generative models for 3D assets
  • Choose and test 3D representations for our use cases
  • Build the pipeline from input to validated asset
  • Write automated geometry validation checks
  • Test outputs in our target renderers and simulators
  • Define evaluation that reflects real usability
  • Work with engineers to fix issues found downstream

What you will bring

  • Shipped or published work on 3D or generative media models
  • Hands-on experience with transformer or diffusion style modelling
  • Clear examples of handling geometry problems in generated output
  • Experience evaluating models beyond visual inspection
  • Comfort working across multimodal inputs such as text and images
  • Ability to explain trade-offs to engineers and product colleagues

Nice to have

  • Experience with simulators or digital twins
  • Background in graphics or rendering
  • Work on controllable or structured generation

Free to use and adapt. If it saves you time, a link back to this page is appreciated.

Skills that matter

  • Multimodal modellingAdvancedCombine information across modalities
  • Generative mediaAdvancedProduce controllable visual, sound or spatial outputs
  • Transformer modellingWorkingImplement attention-based sequence or multimodal models
  • Simulation and digital twinsWorkingBuild useful virtual representations of environments

Other roles in multimodal and generative world models

Hiring nowMultimodal Research ScientistA Multimodal Research Scientist builds models that connect text, images, sound and action, and designs the benchmarks that show where they break…EmergingGenerative Video ResearcherA Generative Video Researcher builds models that produce video that stays consistent over time and responds to control. They make video generation…EmergingWorld Model ResearcherA World Model Researcher builds models that predict how an environment will change, given what has happened and what an agent does next. That gives…

Related roles in other families

Hiring nowPretraining Research EngineerFoundation model pretrainingA pretraining research engineer runs controlled experiments on how large models learn from broad data. The work decides which architectures, data…Hiring nowModel Architecture ResearcherFoundation model pretrainingA Model Architecture Researcher tests changes to how a model is built, to improve its behaviour or efficiency. The work tells a company which…EmergingScaling Science ResearcherFoundation model pretrainingA Scaling Science Researcher works out how model performance changes with compute, data and model size. Their estimates tell a company how to spend…EmergingTraining Dynamics ResearcherFoundation model pretrainingA Training Dynamics Researcher works out why large training runs wobble, stall or diverge, and fixes the cause. Fewer failed runs means less wasted…

How we wrote this

Role definitions are our own editorial work, checked against occupational classifications, standards, research and employer pages. Titles vary by company, and this is a description of the work, not a job advert.

← Multimodal and generative world modelsAll new job families

Start a search

Hiring a Generative 3D Researcher?

Tell us the work. We find the people already doing it, whatever their job title says, and you meet them already interviewed for your role.