AI Engineer

About your missions

YOUR ROLE The AI Engineer leads the design, deployment, and continuous improvement of physical assets to augment or replace manual tasks in semiconductor manufacturing environments.

Key Responsibilities:

1. AI Model Identification & Development

• Analyze technician/operator workflows to identify repetitive, high-precision, or hazardous tasks suitable for AI/physical automation.
• Develop, train, and validate embodied AI models leveraging Vision-Language Models (VLM) and Vision-Language-Action (VLA) frameworks to enable integrated perception, natural language understanding, and task execution capabilities.
• Utilize NVIDIA Omniverse or similar simulation platforms for high-fidelity 3D simulation and digital twin environments to train, test, and optimize physical assets AI models in realistic virtual manufacturing settings.
• Integrate multimodal data (image, sensor, process logs, SPC data) into AI models for robust decision-making and adaptive control.
• Continuously improve models using feedback from simulation and real-world performance.

2. Physical manufacturing asset System Integration

• Collaborate with robotics vendors to adapt physical assets or semiconductor environments (cleanroom compliance, ESD safety, precision control).
• Integrate AI models with physical asset control systems and factory MES/SCADA systems.
• Configure robots/related physical asset for complex task execution such as wafer cassette transfer, equipment start-up/shutdown, parameter verification, inline inspection, and reporting.

3. Simulation and Virtual Training Environment Development

• Design and maintain simulation scenarios within Omniverse/similar platform to replicate manufacturing processes and environments for embodied AI training.
• Employ VLM and VLA techniques to simulate realistic robot-environment interactions, enabling the AI to learn visual perception coupled with language understanding and action planning.
• Validate AI behavior in virtual environments before physical deployment to reduce risk and improve efficiency.

4. Testing, Validation, and Qualification

• Define test plans to validate robot/physical asset accuracy, repeatability, and safety before deployment.
• Establish qualification metrics: task success rate, downtime reduction, human replacement efficiency.
• Work with EHS to ensure compliance with safety and cleanroom standards.

5. Deployment and Continuous Improvement

• Lead pilot projects for physical manufacturing asset-assisted lines and scale up successful deployments.
• Analyze performance data, identify model drift or mechanical degradation, and retrain models as needed.
• Conduct root-cause analysis for failed or abnormal tasks.
• Provide ongoing model tuning and system maintenance.

6. Cross-Functional Collaboration

• Work closely with process engineers, maintenance teams, IT, and data scientists.
• Serve as the liaison between physical manufacturing assets providers and internal automation engineering teams.
• Train human technicians to supervise and interact with physical manufacturing assets systems safely.

About you

YOUR SKILLS & EXPERIENCES

Technical Skills

• Strong understanding of semiconductor manufacturing processes (Front-End and/or Back-End).
• Proficiency in AI/ML model development (Python, TensorFlow/PyTorch, OpenCV).
• Experience with embodied AI frameworks such as Vision-Language Models (VLM) and Vision-Language-Action (VLA) for integrated perception, language understanding, and action modeling.
• Hands-on experience with NVIDIA Omniverse or similar simulation platforms for digital twin and virtual training environments.
• Experience with robotic/humanoid platforms (e.g., Boston Dynamics, UBTech, Hanson Robotics, or custom cobots).
• Knowledge of robot control systems (ROS, PLC interfacing, motion planning).
• Familiar with machine vision systems, defect detection, and sensor fusion.
• Integration of AI models with edge devices and factory systems (OPC-UA, MQTT, MES APIs).
• Experience in cloud technologies is added advantage.

Soft Skills

• Analytical and innovative mindset.
• Strong cross-disciplinary collaboration.
• Clear communication between data science, robotics, and operations teams.
• Strong documentation and safety awareness.

Education & Experience:

• Bachelor's or Master's degree in Robotics, AI/ML, Mechatronics, Electrical Engineering, or Computer Science.
• 3-8 years' experience in semiconductor manufacturing, robotics automation, or AI model deployment.
• Experience deploying AI/vision models in industrial environments (manufacturing, automotive, electronics) is a plus.
• Prior experience with embodied AI training using VLM, VLA, and Omniverse highly desirable.
• This role may require candidate to travel between different manufacturing sites of the company.
• Experience with cleanroom operation or ISO 14644 standards preferred.