AI Engineer

Cat Digital is the digital and technology arm of Caterpillar Inc. responsible for bringing world class digital capabilities to our products and services. With almost one million connected assets worldwide were focused on using IoT and other data technology advanced analytics and AI capabilities to help our customers build a better world.

Build the Digital Backbone of Modern Manufacturing

Were assembling a dynamic team to develop and scale ourManufacturing & Supply Digital Platforma next-generation software framework that transforms how manufacturing and supply operations connect collaborate and optimize.

This platform isnot an ERP system. Its a purpose-built digital layer that integrates data processes and resources across the entire manufacturing lifecyclefrom design and engineering to production and distribution.

Youll be part of a team thats not just building softwarebut shaping the future of how manufacturing works throughAI-driven collaborative and scalable digital solutions.

As part of this initiative youll contribute to:

• System Integration: Seamlessly connecting diverse manufacturing and supply systems data sources and workflows into a unified digital ecosystem.

• Data-Driven Decision Making: Harnessing real-time data collection analysis and visualization to deliver actionable insights and operational intelligence.

• Automation & Optimization: Driving efficiency through intelligent scheduling predictive maintenance and quality controlwithout replacing core transactional systems.

• Enhanced Collaboration: Enabling transparent communication and coordination across teams functions and geographies.

If youre passionate about digital platforms industrial innovation and working with cutting-edge technologiesthis is your opportunity to make a meaningful impact.

Key Responsibilities:

• Design develop and productionize generative AI models (e.g. transformers diffusion GPT BERT) for simulating manufacturing processes and virtual factory environments.

• Build and optimize inference pipelines for digital twin scenarios using OpenUSD and NVIDIA Omniverse deploying models on a variety of edge devices (mobile IoT embedded)

• Leverage and build upon Omniverse/Kit APIs and extensions to enable simulation-data-driven predictive maintenance and process automation.

• Apply quantization pruning compilation and hardware-specific optimizations for low-latency real-time inference in manufacturing use-cases.

• Generate augment and manage large synthetic datasets derived from sensor streams and virtual scenarios for robust model training.

• Conduct performance profiling benchmarking (inference speed accuracy memory) and debugging of AI pipelines within digital twin frameworks.

Skills:

• Deep expertise in Python especially with PyTorch/TensorFlow (model training fine-tuning for vision text and audio; Supervised Unsupervised Reinforcement learning experience is a plus).

• Demonstrated experience in designing modifying developing and implementing AI and Gen AI applications.

• Familiarity with MLOps practices (pipeline automation CI/CD MLflow Docker).

• Knowledge of Machine Learning techniques: anomaly detection predictive analytics manufacturing yields.

Nice to Have Skills:

• Direct experience deploying models with ONNX Runtime TensorRT TFLite.

• Experience with OpenUSD APIs for asset management scene graphs and data exchange between simulation and analytics stacks.

• Experience working with NVIDIA Jetson robotics hardware and/or edge AI accelerators.

• Contribution to open-source projects (USD Omniverse AI libraries).