Principal Perception Architect

Caterpillar Inc.•Irving, TX

4d•$159,120 - $258,570•Onsite

About The Position

We are seeking a Principal Perception Architect / Team Lead to define and lead the end-to-end perception architecture for advanced autonomous machine systems. This role will be responsible for shaping how machines sense, understand, represent, validate, and act on the external world in real operating environments. The Principal Perception Architect will serve as the senior technical leader for the Perception Architecture Team, owning the overall perception system architecture across sensor integration, sensor fusion, machine learning-based perception, scene understanding, world model inputs, perception data flow, validation, and safety analysis. This person will ensure that perception capabilities are not developed as isolated algorithms, but as part of a complete autonomy architecture that supports real-time machine behavior, edge deployment, simulation, validation, and safe operation. This role requires a strong systems thinker who can connect perception technology with actual machine autonomy needs. The ideal candidate will understand how cameras, lidar, radar, stereo, ultrasonic, GNSS, machine state data, perception models, world models, and edge compute platforms come together to support autonomous operation. They must be comfortable making complex architecture tradeoffs across perception accuracy, latency, compute, cost, sensor coverage, robustness, safety, scalability, and field reliability. The Principal Perception Architect will work closely with the Robotics team to ensure perception outputs are useful for planning, control, and machine behavior. They will also partner with Edge Architecture, Machine Platform, Simulation, Validation, Safety, and Product teams to ensure perception outputs are deployable, replayable, testable, traceable, and ready for real-world autonomy use. A major part of this role is defining clear technical contracts between perception and downstream consumers, including world model services, robotics autonomy logic, simulation systems, validation pipelines, and edge runtime platforms. This position is both a deep technical architecture role and a team leadership role. The Principal Perception Architect will guide other architects across sensor integration and fusion, perception machine learning, perception data and simulation, and perception validation and safety analysis. They will be expected to set direction, drive architecture reviews, resolve cross-domain technical conflicts, identify system-level risks, and create a practical roadmap that moves perception capabilities from concept to machine-ready autonomy.

Requirements

Analytical Thinking: Expert Knowledge of techniques and tools that promote effective analysis; ability to determine the root cause of organizational problems and create alternative solutions that resolve these problems.
Platform Architecture: Expert knowledge of technologies and methods to design processing mechanisms and roadmaps to execute business application systems; ability to design these roadmaps and deploy supportive interfaces for end-users to access related systems, in accordance with standards and processes.
Data Architecture: Expert knowledge of processes, techniques and factors that affect data architecture; ability to design blueprints on how to integrate data resources for business processes and functional support. Ability to utilize design methodologies, tools and techniques to convert business requirements and logical models into a technical design.
Effective Communications: Expert understanding of effective communication concepts, tools and techniques; ability to effectively transmit, receive, and accurately interpret ideas, information, and needs through the application of appropriate communication behaviors.
Organizational Leadership: Expert knowledge of leadership concepts and ability to use strategies and skills to enlist others in setting, embracing and achieving objectives.
Progressive career in software engineering and architecture with a focus on data (typically 10 years+)
Demonstratable experience leading technology development for enterprise scale systems in public cloud and hybrid eco-systems Team leadership (typically 8+ years)
Understanding of C++, python programming experience

Nice To Haves

Ability to take ownership of end-to-end perception architecture
Strong autonomy systems thinking with ability to connect perception outputs to robotics, planning, control, simulation, edge runtime, machine platform interfaces, safety analysis, and field validation.
Has experience with deployed autonomous, robotic, vehicle, industrial, agriculture, construction, mining, or mobile machine systems.
Multi-sensor integration and fusion depth with strong understanding of camera, lidar, radar, stereo, ultrasonic, GNSS, localization inputs, and machine state data.
Robotics-facing and edge-aware mindset:
Validation, safety, Physical AI platform, and NVIDIA ecosystem familiarity:
Perception Architecture Leadership & Engineering Execution:

Responsibilities

Define and own the end-to-end perception architecture for autonomous machine systems, including sensor integration, sensor fusion, perception ML, scene understanding, world model inputs, data flow, validation, and safety analysis.
Lead the Perception Architecture Team by setting technical direction, aligning architecture priorities, guiding key design decisions, and ensuring the team works as one cohesive architecture function.
Create the perception technical roadmap that connects near-term product needs with long-term autonomy capability, scalability, reliability, and platform reuse.
Drive architecture decisions and tradeoffs across perception accuracy, latency, compute usage, sensor coverage, cost, power, robustness, safety, and deployability on edge platforms.
Ensure perception architecture supports real-world operating conditions, including sensor degradation, harsh environments, blind spots, poor visibility, machine vibration, calibration drift, compute limits, and uncertain or incomplete data.
Partner cross-functionally with Edge Architecture, Simulation, Data, Machine Platform, Safety, Validation, Product, and Robotics teams to ensure perception capabilities are deployable, replayable, traceable, testable, and ready for real-world autonomy use.