Beyond Medicine is an AI-based digital therapeutic development startup developing the world’s first digital therapy for temporomandibular disorders (TMD) called ‘Clickless’. This position is responsible for camera-based image data processing and analysis, development and enhancement of computer vision models, and implementing self-/semi-supervised and domain adaptation strategies to efficiently train small labeled and large unlabeled medical data.
- Preprocessing of camera image and video-based temporomandibular joint (TMJ) imaging data and design/establishment of learning pipeline
- Development and enhancement of computer vision models for the TMD digital therapeutic pipeline
- Establishment and implementation of self-/semi-supervised learning and domain adaptation strategies suitable for small-label/mass-unlabeled medical data
- Master's degree or higher in computer vision/machine learning or over 3 years of relevant practical experience
- Proficiency in Python and PyTorch/TensorFlow
- Experience in developing image classification, detection, segmentation models
- Ability to lead problem definition and resolution in a startup environment and facilitate smooth team communication
- Experience in the establishment and operation of MLOps (MLflow, CI/CD, automated monitoring)
- Publication or presentation of papers at international conferences/journals such as CVPR, MICCAI, or competition win experience
- Experience utilizing limited label data such as transfer learning, self-/semi-supervised learning
- Experience in developing or commercializing face recognition models
Benefits and Work Environment
- Flexible working hours
- Support for the latest development equipment and research environments
- Provision of snacks and beverages in the office
Document screening → 1st interview (practical) → 2nd interview (executive) → final offer
Submitted documents: Resume, portfolio (including project details and links to code repositories is a plus)
