Computer Vision Engineer

AI-powered laser-weeding systems

Computer Vision Engineer

What you'll do

• Design, train, and optimise CV/ML algorithms for

  weed detection, crop segmentation, and health classification

  .
• Build and refine

  data pipelines

  : labeling, augmentation, dataset versioning, and quality control.
• Improve

  accuracy, robustness, and FPS

  across varied farm conditions (lighting, soil, rains, dust).
• Deploy and optimise models on

  edge hardware

  (Jetson, Qualcomm, x86, custom accelerators) with tight compute budgets.
• Implement

  classical CV + deep learning

  (OpenCV, geometric transforms, segmentation/detection networks).
• Build and maintain

  real-time inference pipelines

  using

  GStreamer, DeepStream

  , CUDA plugins, and custom kernels.
• Convert and optimise models using

  TensorRT, ONNX Runtime

  , quantisation (INT8/FP16), and graph optimisation.
• Work with the perception team on

  NIR/RGB fusion

  , extrinsic calibration, and sensor synchronisation.
• Analyse field failures and iterate on models

  fast

  .
• Build internal tools for

  evaluation, visualization, and automated benchmarking

  .

What you need

• Strong grounding in

  computer vision, machine learning, and image processing

  .
• Experience with

  PyTorch/TensorFlow

  , ONNX, TensorRT, or similar optimisation stacks.
• Comfort with

  C++/Python

  , CUDA, and real-time edge deployments.
• Hands-on experience with

  GStreamer

  and/or

  NVIDIA DeepStream

  pipelines.
• Solid understanding of model evaluation, metric design, class imbalance, and dataset bias.
• Ability to work with large, messy, real-world datasets.
• Experience withembedded GPU optimisation, Jetson NX/Orin deployments, or Qualcomm accelerators.
• Familiarity with robotics perception, ROS, or sensor integration.

when, where, and how our robot fires a laser at weeds