Principal Software Engineer (nvidia Platform)

Caterpillar Caterpillar · Industrial · Irving, TX

Principal Software Engineer to lead architecture and integration for Physical AI/Digital Twin, simulation platforms, and cloud-native services. Focus on defining standards, leading system design, and enabling teams to deliver secure, scalable platform capabilities. Will also lead adoption of AI-assisted and agentic development workflows.

What you'd actually do

  1. Define and drive end-to-end architecture and integration strategy for simulation platforms, digital twin systems, and backend microservices APIs
  2. Lead design and delivery of scalable cloud-native services, integration adapters, and data/asset exchange systems
  3. Establish and enforce design patterns, service contracts, and integration standards (REST, gRPC, event-driven) across distributed systems
  4. Provide technical leadership on complex system design and platform scalability challenges
  5. Lead adoption of AI-assisted and agentic development workflows, defining standards for prompt quality, validation, and code generation practices

Skills

Required

  • Software Product Design/Architecture
  • Software Product Technical Knowledge
  • Decision Making and Critical Thinking
  • Effective Communications
  • cloud-native microservices
  • distributed systems
  • APIs (REST, gRPC, event-driven)
  • Python
  • Java
  • responsible AI practices
  • secure software development

Nice to have

  • simulation
  • NVIDIA ecosystem
  • robotics
  • digital twin
  • industrial automation platforms
  • integration architectures
  • observability patterns
  • AI-assisted/agentic development workflows
  • validating AI-generated code
  • building production AI agents
  • agent orchestration frameworks
  • LangChain
  • LangGraph
  • integration LLMs with external tools, APIs, databases, and retrieval systems
  • designing evals, guardrails, and monitoring for agent reliability
  • Multi-agent workflow design
  • memory/context management
  • Prompt optimization
  • latency reduction
  • cost control
  • MLOps
  • model deployment

What the JD emphasized

  • Physical AI / Digital Twin Integration
  • AI-assisted and agentic development workflows
  • production AI agents
  • agent orchestration frameworks
  • integration LLMs with external tools, APIs, databases, and retrieval systems
  • designing evals, guardrails, and monitoring for agent reliability
  • Multi-agent workflow design and memory/context management

Other signals

  • Physical AI / Digital Twin Integration
  • AI-assisted and agentic development workflows
  • production AI agents