AI Engineer - Computer Vision

Title : AI Engineer with Computer VisionRoles & ResponsibilitiesTrain and fine-tune instance segmentation and object detection models (Mask R-CNN, YOLO-family) to detect irrigation and landscape symbols across mixed symbol libraries, varying scales and rotationsBuild the hatch and bed-line segmentation pipeline that measures areas for sod, seed, mulch, paving and similar, including cases where identical hatches carry different labels or appear nowhere in the legendDesign the tiling and preprocessing strategy for large-format plan sheets, and own the trade-off between tile count, detection recall and inference costImplement calibrated confidence scoring, and the "unknown — needs review” path so unfamiliar symbols surface instead of being misclassified with false confidenceRun the three fine-tuning rounds: train, evaluate against acceptance criteria, error-analyse which classes underperform, and carry a written remediation plan into the next roundDeploy models to Amazon SageMaker asynchronous endpoints with scale-to-zero, and keep releases immutable and version-tagged so any release can be rolled backWork with the QA engineer on per-class acceptance measurement, no-regression checks and the calibration and held-out splitSkills requiredStrong Python, and production experience with PyTorch on detection or segmentation tasks — not just notebook experimentsHands-on work with Mask R-CNN, YOLO, Detectron2 or similar, including training on custom datasets and diagnosing class-level failureOpenCV and image preprocessing: tiling, rotation, scale normalisation, morphologyPDF and raster handling at scale (PyMuPDF, pdf2image, Poppler, Pillow)A working understanding of detection metrics — mAP, IoU, precision and recall per class — and why aggregate accuracy hides per-class regressionAWS: S3, SQS, and either SageMaker or comparable managed inferenceNice to haveExperience reading engineering, architectural or CAD-derived drawingsActive learning or human-in-the-loop correction pipelinesModel optimisation for inference: TensorRT, ONNX, quantisationAnnotation tooling and labelling workflow design