Genesis-World: Core Physics Engineer

Genesis•San Carlos, CA

12d

About The Position

Genesis-World is an open-source, general-purpose simulation platform for physical AI from Genesis AI. It features a unified multi-physics engine supporting rigid bodies, FEM, MPM, particles, cloth, and fluids. The platform includes an in-house renderer, Nyx, for real-time photorealistic rendering, and supports various sensors like cameras, lidar, IMU, and contact forces, along with advanced tactile simulation. Genesis-World is developing a comprehensive and fast Incremental Potential Contact (IPC) solver. The platform powers real business applications and is designed for Python-first operation across different hardware backends (CUDA, AMD ROCm, Apple Metal, Vulkan, x86, and ARM64). It aims to accelerate robotics development by providing a simulation environment that runs significantly faster than real-world hardware and correlates well with on-hardware results, enabling physical AI to improve at the speed of compute. The Core Physics Engineer will advance the physics capabilities of Genesis-World, focusing on shipping production-ready simulation features crucial for the company's internal needs. This role involves end-to-end research, from algorithm development to delivering tested, documented code for robot-learning pipelines. The core responsibilities include improving the engine's speed, completeness (covering a wide range of physics phenomena), fidelity (realistic models), versatility (extensible multi-physics), and scalability (from workstations to city scale). The goal is to establish Genesis-World as the premier simulator for physical AI.

Requirements

A strong background in physics-based simulation, preferably related to robotics.
Experience with RBD, FEM, MPM, SPH, IPC, XPBD, VBD, ABD.
Expertise in constrained optimization and numerical integration of stiff systems.
A track record of shipping simulation code that others rely on (in an engine, industry, or research codebase).
Solid HPC programming skills (CPU and/or GPU).
An instinct for what makes a numerical method fast in practice.
Rigor in validation: analytical closed forms, cross-engine consistency, real-world data.

Nice To Haves

Publications in simulation, graphics, or robotics venues (SIGGRAPH, ICRA, IROS, CoRL, RSS).
Contributions to an open-source physics engine.

Responsibilities

Ship production-ready simulation capabilities that matter for the company's internal needs.
Conduct end-to-end research, from algorithm to merged, tested, documented code.
Improve the engine's speed through faster algorithms and selective fidelity.
Expand the engine's completeness to cover a wide range of physics phenomena (water, air flow, gravel, tendons, organs, etc.).
Enhance the fidelity of physics models, including contact, friction, deformation, energy, actuation, and materials.
Develop versatile and extensible multi-physics capabilities with two-way, constraint-based coupling between solvers.
Ensure scalability of the engine from workstations to factory and city scales.
Invent physics level-of-detail (LOD) to simulate relevant interactions at full fidelity and others coarsely.
Develop solutions for heterogeneous environments where each parallel world can hold a different model.
Implement adaptive timesteps per island using error-based control and Time-of-Impact stepping.
Couple heterogeneous grey-box solvers efficiently and accurately using state-of-the-art methods like ADMM.
Improve the scalability of rigid constraint solvers.
Unify contact resolution, including hydro-elastic compliance, unilateral constraints, and sequential impulses.
Handle closed kinematic loops intrinsically for numerical stability and speed.
Write physics in plain Python, leveraging Quadrants for performance on various backends.
Validate physics implementations through analytical closed forms, other engines, and real-world data.