Principal Architect – Simulation Engine (Physical AI)

Caterpillar Inc.•Irving, TX

8d•$159,120 - $258,570•Onsite

About The Position

The Principal Digital Architect – Simulation Engine defines the technical vision, architecture, and roadmap for the simulation platform supporting Physical AI and autonomy development for Caterpillar construction and mining machines. The platform must simulate machine and material physics, sensors operating in varied environmental conditions, and dynamic jobsites. It must deliver deterministic execution, high fidelity where needed, and scale to millions of scenario executions per day across GPU clusters. The architect sets direction for core simulation abstractions, including world and asset representation, physics and rendering, sensor models, scenario definition and generation, open-loop replay, closed-loop software-in-the-loop and hardware-in-the-loop execution, and interfaces for reinforcement learning and neural simulation. The role establishes measurable standards for fidelity, determinism, performance, and developer experience, and leads sim-to-real correlation using field data. Working with simulation engineering, Digital Twin and AI Factory architects, autonomy and perception teams, and controls and machine engineering, the architect helps converge on a unified platform, reduce fragmentation, and improve engineering velocity. The role also develops architecture deliverables, designs application infrastructure, evaluates and deploys technologies, and collaborates across teams to deliver digital solutions aligned with business requirements.

Requirements

Knowledge of requirements-analysis tools, methods, and techniques, and ability to elicit, analyze, and record business functionality and non-functional requirements.
Knowledge of analytical techniques and tools, and ability to identify root causes and develop alternative solutions.
Knowledge of platform architecture technologies and methods, with ability to design processing mechanisms, roadmaps, and end-user interfaces in accordance with standards and processes.
Knowledge of target architecture, with ability to develop IT blueprints and roadmaps aligned with business strategies and objectives.
Effective communication skills, including the ability to transmit, receive, and interpret ideas, information, and needs.
Bachelor’s, master’s, or doctoral degree in Computer Science, Robotics, Mechanical Engineering, Physics, or a related field, or equivalent experience.

Nice To Haves

Typically 8+ years in software engineering, including ownership of a production simulation platform for autonomous vehicles or robotics, covering core abstractions, deterministic execution, replay and closed-loop modes, and SIL/HIL integration.
Hands-on experience with NVIDIA Omniverse / Isaac Sim (Kit, OpenUSD, PhysX, RTX) and/or Unreal Engine; Applied Intuition, CARLA, Isaac Lab, MuJoCo, or Project Chrono experience is a plus.
Experience with multibody dynamics, contact and friction, deformable terrain and granular material, track/tire-ground interaction, hydraulics, and powertrain; models calibrated and validated against test data.
Experience with physically based camera, LiDAR, radar, GNSS, and IMU models, including dust, weather, noise, and calibration effects, validated against real sensor logs; exposure to 3D Gaussian Splatting / NeRF.
Experience with large outdoor terrain and geospatial environments and procedural or generative scenario generation with coverage tied to the ODD; familiarity with ASAM OpenSCENARIO and SOTIF / ISO 26262.
Experience with Gym-style reinforcement-learning environments and large-scale closed-loop training (Isaac Lab, Ray), and understanding of world foundation models such as NVIDIA Cosmos alongside physics simulation.
Experience running simulation on GPU clusters and cloud using Kubernetes, Slurm, Ray, or CUDA, with demonstrated improvement in throughput, determinism, and cost per simulation.
Expertise in modern C++ and strong Python skills; experience with gRPC / Protocol Buffers, Bazel or CMake, and CI/CD for large monorepos; Rust or CUDA is a plus.
Experience serving as an architectural authority across multiple teams, influencing senior leaders, and mentoring principal, staff, and senior engineers.
Domain experience in off-highway, mining, or construction autonomy, or AV/robotics simulation at companies such as Applied Intuition, NVIDIA, Waabi, Wayve, Zoox, GM/Cruise, or Foretellix.

Responsibilities

Define the technical vision, architecture, and roadmap for Caterpillar’s simulation platform for Physical AI and autonomy.
Set direction for simulation abstractions covering world and asset representation, physics and rendering, sensors, scenarios, execution modes, and training interfaces.
Establish measurable standards for simulation fidelity, determinism, performance, and developer experience.
Lead the sim-to-real correlation program to validate simulator trustworthiness against field data.
Develop detailed architecture deliverables to solve business problems.
Design application technical infrastructure, including databases, programming languages, utilities, and testing approaches.
Lead evaluation and deployment of technologies that add or enhance digital technical capabilities.
Address application business requirements and collaborate with cross-functional teams to deliver digital solutions and business results.
Work with simulation engineering, Digital Twin and AI Factory architects, autonomy and perception teams, and controls and machine engineering to converge on a platform, reduce fragmentation, and improve engineering velocity.