Principal Architect – Digital Twin

Caterpillar Inc.•Irving, TX

2d•$147,760 - $240,110•Onsite

About The Position

The Principal Architect – Digital Twin is responsible for the architecture of the digital twins used to develop and validate Physical AI, simulation, and autonomy. The role covers machine twins, jobsite and environment twins, and sensor twins. Each twin must be simulation-ready, physically based, versioned, and validated against real machines and sites. The architect defines the asset and scene pipeline that converts CAD, PLM, survey, and reality-capture data into simulation-ready OpenUSD assets and environments for NVIDIA Omniverse / Isaac Sim and Unreal Engine. The role also defines data models connecting twins to machine telemetry and fleet data, fidelity validation and correlation programs, and synthetic-data and neural-reconstruction workflows. The position works closely with the Simulation Engine architect and the AI Factory architect. The role bridges architecture and engineering by translating the digital twin vision into detailed solution designs, reusable patterns, standards, and implementation guidance, and by guiding the engineering teams that build them. This position is based onsite at Caterpillar’s Irving, Texas office, five days per week. Domestic relocation assistance and visa sponsorship are available.

Requirements

Knowledge of requirements-analysis tools, methods, and techniques, with the ability to elicit, analyze, and record business functionality and non-functionality requirements.
Analytical thinking skills, including the ability to identify root causes and develop alternative solutions.
Knowledge of platform architecture and the ability to design processing mechanisms, roadmaps, and supportive interfaces in accordance with standards and processes.
Knowledge of target architecture and the ability to develop an IT blueprint and roadmap aligned with business strategies and objectives.
Effective communication skills, including the ability to transmit, receive, and accurately interpret ideas, information, and needs.
Bachelor’s degree in Computer Science, Robotics, Mechanical Engineering, Physics, or a related field.
Ability to work full-time onsite at the Irving, TX office, five days a week.

Nice To Haves

Master’s or doctoral degree preferred.
Experience delivering physically based digital twins of machines, vehicles, robots, or industrial sites for simulation, validation, or AI training, including fidelity tiers, versioning, and lifecycle.
Hands-on experience with NVIDIA Omniverse (Kit, Isaac Sim, Replicator, Nucleus) and OpenUSD, including composition, schemas, USD Physics, SimReady assets, and CAD-to-USD conversion at scale.
Experience with Unreal Engine or Unity for large outdoor environments.
Experience with multibody dynamics, articulations, hydraulics, powertrains, and implement-material interaction using URDF, MJCF, or USD Physics; model calibration and system identification against measured data.
CAE/PLM exposure, such as Siemens NX, Ansys, Modelica, or Adams.
Experience building 3D environments from drone photogrammetry, LiDAR, survey and design surfaces, and GIS data, and with procedural site generation.
Construction or mining site data experience, such as Trimble, Topcon, Hexagon, or Cat MineStar / Grade.
Experience with physically based camera, LiDAR, radar, GNSS, and IMU models, calibration frames, environmental effects, and synthetic data with ground truth and domain randomization.
Experience with 3D Gaussian Splatting, NeRF, or world foundation models such as NVIDIA Cosmos for neural reconstruction and real-to-sim workflows.
Experience linking twins to live machine telemetry and fleet data using ROS 2, MQTT, Kafka, DDS, or edge-to-cloud pipelines.
Experience establishing sim-to-real acceptance criteria, test-data correlation, scenario-based and regression testing, HIL/SIL correlation, and quantified sim-to-real gap metrics.
Strong Python and C++ skills; experience with GPU/CUDA workloads, containers and Kubernetes, CI/CD, and AWS or Azure; comfort with AI coding tools.
Architecture leadership translating architecture into implementation plans, reusable patterns, and standards across teams, and providing technical leadership to engineers.
Domain exposure to off-highway, construction, mining, or agriculture machines, or digital twin programs at NVIDIA, Applied Intuition, Volvo Autonomous Solutions, Komatsu, John Deere, Siemens, or Duality AI.

Responsibilities

Develop detailed architecture deliverables to solve business problems.
Design application technical infrastructure, including databases, programming languages, utilities, and testing approaches.
Lead the evaluation and deployment of new technologies to add or enhance digital technical capabilities.
Address business requirements for applications and collaborate with cross-functional teams to deliver digital solutions and business results.
Define architecture for machine, jobsite/environment, and sensor digital twins used in Physical AI, simulation, and autonomy.
Define asset and scene pipelines that convert CAD, PLM, survey, and reality-capture data into simulation-ready OpenUSD assets and environments.
Define data models connecting twins to machine telemetry and fleet data so twins reflect as-operated machines and as-built sites.
Establish validation and correlation programs, including test-data correlation, scenario-based testing, regression testing, and telematics comparison, to quantify twin fidelity.
Develop synthetic-data and neural-reconstruction workflows to scale twin creation.
Translate the digital twin vision into solution designs, reusable patterns, standards, implementation guidance, and engineering plans.
Guide engineering teams building digital twin solutions and collaborate with the Simulation Engine and AI Factory architects.