SPX Technologies is looking to develop AI models and automation to increase the accuracy, speed, and ease of inspections for their CUES business, which provides solutions for water, wastewater, and storm water infrastructure systems.
Requirements
- 5+ of professional software experience, including 3+ years focused on deep learning / computer vision.
- Proven record of shipping CV/ML models into production on robotic, embedded, or edge‑AI platforms.
- Hands‑on experience with object detection and segmentation architectures (YOLO, DETR, EfficientDet, etc.).
- Proficiency in Python and comfort with at least one systems language (e.g., C++, Rust)
- Fluency in Linux development, Git‑based workflows, CMake, and containerization (Docker/Podman).
- Strong understanding of camera physics, image formation, stereo geometry, and depth estimation.
- Experience with 3‑D reconstruction, point‑cloud processing (PCL, Open3D), or LiDAR fusion.
Responsibilities
- Design, implement, and optimize deep‑learning models using tools such as YOLOv_, Faster R‑CNN, Mask R‑CNN, PyTorch, for multi‑class defect detection, segmentation, and severity grading of imagery.
- Curate, Manage, and augment large video / image datasets; define labeling guidelines and manage annotation workflows (Labelbox, etc.).
- Design custom metrics to evaluate model performance in the field
- Tune hyper‑parameters, perform ablation studies, and apply automated model‑selection / NAS techniques to maximize precision‑recall under real‑time constraints.
- Convert and optimize models for embedded inference using tool such as TensorRT, ONNX, NVIDIA Jetson, Intel OpenVINO.
- Build reproducible training / CI‑CD pipelines with Docker, GitHub Actions, PyTorch Lightning, and MLflow; automate experiment tracking and model registry.
- Collaborate with robotics engineers to integrate models into the real‑world systems.
Other
- Excellent verbal and written communication skills in English; able to convey complex ML concepts to cross‑functional audiences.
- Ability to travel occasionally (domestic and international) to test sites and customer locations.
- Travel up to 10%
- Hybrid Role, Minimum 3 days in office per week.
- Bachelor’s in Computer Science, Computer Engineering, or Electrical Engineering required.
