Senior Edge AI Engineer, Industrial Inspection

The challenge

The next-gen inspection system needs to detect 12 defect types on PCB assemblies at 300 parts per minute with <0.1% false reject rate. The target platform is an embedded NVIDIA Jetson Orin at 1/4 the cost of current systems.

Role overview

You develop and optimize deep learning models for real-time visual inspection on embedded hardware. From labeled defect datasets to trained and quantized models deployed on edge compute platforms in factory environments -- the full pipeline from data to production inference.

Specializations

Technical requirements

• MS in Computer Science, Computer Vision, or Machine Learning
• 5+ years developing computer vision models for industrial or manufacturing applications
• Experience with model optimization techniques: quantization (INT8, FP16), pruning, knowledge distillation
• Proficiency with PyTorch and NVIDIA TensorRT for model training and deployment
• Experience with object detection architectures (YOLO, SSD, Faster R-CNN) and instance segmentation
• Familiarity with machine vision hardware (industrial cameras, lighting, optics)

Key responsibilities

• Develop and train defect detection models for visual inspection of manufactured parts
• Optimize models for deployment on embedded platforms (NVIDIA Jetson, Intel NCS) using quantization and pruning
• Design data labeling workflows and quality assurance processes for manufacturing defect datasets
• Implement inference pipelines that achieve <50ms latency per inspection on edge hardware
• Collaborate with application engineers to integrate models into production inspection systems
• Monitor deployed model performance and implement continuous learning pipelines

Compensation & Benefits

• Base salary: $175,000 - $220,000
• Stock options (Series C company)
• Annual bonus 8-12%
• Hybrid work (3 days on-site near customer labs)
• Professional development budget ($4,000)