Gaussian Splatting Jobs (NOW HIRING)

We are looking for a Software Development Engineer with deep expertise in 3D data, geometry processing, physics literacy and asset pipelines to join the Robotics Simulation team at Amazon Robotics. This role focuses on building scalable tooling and workflows that produce simulation-ready 3D content with high visual and physics fidelity.
The day-to-day work is grounded in traditional 3D engineering: CAD conversion, geometry optimization, USD authoring, format interop, and pipeline hardening. This is roughly 70-80% of the role

The remaining 20-30% involves integrating emerging techniques (Gaussian Splatting, neural reconstruction, generative 3D) into production workflows as the field matures. We need someone who excels at the fundamentals and can bring modern approaches online as they become production-viable.
At this level, this role additionally carries ownership of the team's 3D pipeline architecture, drives cross-team technical standards, mentors junior engineers on geometry processing and USD best practices, and influences the broader simulation asset strategy across multiple robotics programs.
Key job responsibilities
* Design and harden automated pipelines that convert CAD assemblies into simulation-ready USD assets with correct articulation, collision geometry, material properties, and physics parameters.
* Implement geometry processing tools (mesh optimization, convex decomposition, feature culling, LOD generation) balancing fidelity with simulation performance.
* Integrate emerging 3D techniques (Gaussian Splatting, neural reconstruction, generative 3D, LLM-orchestrated scene building) into production workflows as they mature.
* Define and enforce 3D asset standards and build validation tooling that programmatically checks assets against quality gates.
* Collaborate closely with Applied Scientists to ensure assets meet downstream training pipeline requirements, iterating on collision geometry and domain randomization parameters based on policy performance feedback.
* Own the architectural direction of the team's 3D pipeline ecosystem, driving the roadmap and long-term technology choices.
* Mentor junior SDEs and Technical Artists on geometry processing, USD best practices, and pipeline engineering.
* Author design documents and lead technical reviews; drive cross-team alignment on asset standards with partner organizations.
A day in the life
Amazon offers a full range of benefits that support you and eligible family members, including domestic partners and their children. Benefits can vary by location, the number of regularly scheduled hours you work, length of employment, and job status such as seasonal or temporary employment

The benefits that generally apply to regular, full-time employees include:
1. Medical, Dental, and Vision Coverage
2. Maternity and Parental Leave Options
3.

Paid Time Off (PTO)
4. 401(k) Plan
If you are not sure that every qualification on the list above describes you exactly, we'd still love to hear from you. At Amazon, we value people with unique backgrounds, experiences, and skillsets.

If you're passionate about this role and want to make an impact on a global scale, please apply!
About the team
The Robotics Simulation team is a ~30-person multidisciplinary organization of SDEs, Applied Scientists, and Technical Artists at Amazon Robotics. We build the simulation infrastructure that powers Physical AI development, from photorealistic synthetic data to GPU-accelerated training environments. Our tools enable robots to be designed, trained, and validated entirely in simulation before physical hardware exists, compressing development timelines and de-risking hardware programs across Amazon Robotics.

This role is what makes that possible: without simulation-ready 3D content, nothing downstream (policy training, synthetic data, sim-to-real transfer) can function.
This role reports into the Robotics Simulation organization and works directly alongside Applied Scientists building robot learning pipelines. Your assets and tools are the foundation their training runs on: when collision geometry is wrong, policies fail; when visual fidelity is off, sim-to-real transfer breaks. That tight feedback loop between 3D engineering and science is a defining characteristic of the role.