Staff/Principal Full-Stack AI Engineer, Robotic and Digital Twin_Taoyuan / Taichung

Apply here: https://micron.wd1.myworkdayjobs.com/zh-TW/External/job/Taichung---Fab-16-Taiwan/SR-Full-Stack-AI-Engineer--IPAI--SMAI_JR106378

Job Responsibilities

• Develop responsive front-ends in React, Angular, or Streamlit, backed by Python/FastAPI, with strong type-safety and test coverage
• Design and develop full-stack integration of robotics platforms, Autonomous Mobile Robots (AMRs), tool automation, and control systems with smart manufacturing applications and the Micron smart manufacturing ecosystem
• Design and implement digital twin environments — constructing 3D representations of fab layouts, AMR fleets, and tool stations using Gazebo, NVIDIA Omniverse, or Unity Robotics Hub for simulation, path planning validation, and operational rehearsal
• Interface with existing Smart Manufacturing systems to feed structured data into AI pipelines and 3D digital twins
• Integrate LLMs and agentic AI frameworks into production workflows for fab engineers
• Build AI-assisted robotics orchestration and develop 3D-aware AI features — enabling models to reason over spatial data (equipment positions, robot trajectories, fab zone maps) for anomaly detection, route optimization, and predictive maintenance
• Containerize services with Docker, deploy via Kubernetes / OpenShift, and automate releases with CI/CD
• Package and deploy robotics software stacks and 3D rendering services as containerized microservices, supporting both edge (fab-floor) and cloud execution environments
• Manage GPU-accelerated inference workloads for both AI models and real-time 3D physics simulation (e.g., NVIDIA Isaac Sim on cloud)
• Optimize performance and cost: caching, batching, GPU utilization, model quantization, and right-sizing inference for edge or cloud
• Apply Responsible AI practices: input validation, prompt-injection defense, PII/IP redaction, model risk and audit logging
• Validate 3D simulation scenarios against real-world operational constraints before deployment to ensure zero-defect autonomous operation

Qualifications

• Bachelor's or Master's degree in Computer Science, Software Engineering, Data Science, or a related field; equivalent industry experience accepted
• Track record of delivering in complex, cross-functional environments — collaborating across data science, robotics, operations, and infrastructure teams
• 5+ years building and shipping production web or AI applications end to end
• Hands-on experience with at least one cloud (AWS, Azure, or GCP) and with Docker + Kubernetes; CI/CD pipeline ownership
• Experience designing data ingestion and feature pipelines for AI/ML workloads at scale
• Working knowledge of LLMs and the GenAI stack: prompt engineering, function/tool calling, RAG, agentic patterns; familiarity with MLOps / LLMOps and model evaluation/safety practices
• Hands-on with digital twin platforms: NVIDIA Omniverse, Unity Robotics Hub, Gazebo, Siemens Tecnomatix, or Unreal Engine would be an advantage
• Proven experience in robotics development using ROS/ROS 2 frameworks and Isaac Sim/Lab, with hands-on expertise in SLAM perception systems, LiDAR/depth camera data processing, and sim-to-real transfer techniques such as imitation learning and domain randomization.
• Proven experience in integration of SCADA/FMCS/BMS systems for smart facilities and EHS , ideally in semiconductor, data center, pharmaceutical, or other mission-critical/cleanroom environments
• Deep proficiency in the Autodesk ecosystem (Revit, Navisworks, ACC, ReCap), coupled with expertise in point cloud processing, photogrammetry, Gaussian splatting, and geometry optimization via IFC and Revit APIs.Deep proficiency in the Autodesk ecosystem (Revit, Navisworks, ACC, ReCap), coupled with expertise in point cloud processing, photogrammetry, Gaussian splatting, and geometry optimization via IFC and Revit APIs.