About this role
Responsibilities:
- Develop and fine-tune computer-vision models (detection, tracking, segmentation) using PyTorch/TensorFlow and OpenCV
- Work hands-on with YOLO, RF-DETR and Vision-Language Models, fine-tuning for project use cases
- Curate, preprocess and augment image/video datasets, maintaining a feedback loop with the annotation team
- Train, tune and evaluate models against metrics (mAP, precision, recall); track experiments via MLflow/Weights & Biases
- Apply model-optimisation techniques (TensorRT, ONNX) with deployment constraints in mind
Qualifications:
- Bachelor's or Master's degree in Computer Science, Data Science, Machine Learning or related field
- 3-7 years of applied machine-learning experience with a computer-vision focus
- Strong Python with hands-on PyTorch or TensorFlow and OpenCV
- Demonstrated experience with detection, tracking and segmentation models (e.g. YOLO, RF-DETR)
- Experience fine-tuning Vision-Language Models and familiarity with experiment-tracking tools is an advantage