AI Engineer

AI Engineer Computer Vision & OCT Imaging ScaleiQ
· Digital Health / Machine Learning About the engagement

ScaleiQ is developing a hybrid quantum approach to early disease detection, correlating biomarkers with optical coherence tomography (OCT) scans to surface faint, high-dimensional patterns that precede the clinical diagnosis of conditions like cancer and Alzheimer’s. We’re seeking an AI Engineer on a contract basis with deep computer vision experience — specifically with OCT imaging and high-dimensional, longitudinal data — to help build and train the models at the core of this work.

This is a contract/consulting role with the potential to grow as the company moves toward its seed round.

• Design, train, and optimize convolutional neural networks (CNNs) and related architectures for medical image analysis, with a focus on OCT scans.
• Work with high-dimensional, longitudinal imaging data — handling temporal sequences, patient-level tracking, and change detection over time.
• Build reusable libraries, pipelines, and tooling for data preprocessing, model training, and evaluation at scale.
• Develop training strategies capable of handling very large OCT datasets (potentially millions of images), including data augmentation, distributed training, and efficient handling of class imbalance and label scarcity.
• Collaborate with our scientific and quantum teams to integrate classical CV pipelines with hybrid quantum methods, applying each where it performs best.
• Contribute to reproducible benchmarking and rigorous evaluation against classical baselines.

• Strong background in deep learning for computer vision, with hands-on experience building and training CNNs.
• Direct experience with medical imaging — OCT specifically is strongly preferred; related modalities (retinal imaging, MRI, CT, histopathology) are valued.
• Experience working with high-dimensional and/or longitudinal datasets.
• Proficiency in Python and modern ML frameworks (PyTorch, Tensor Flow, or similar).
• Track record of building libraries, reusable components, or production-grade training pipelines.
• Experience training models on large-scale image datasets and a clear point of view on how you’d approach training on millions of OCT images.
• Bonus: familiarity with self-supervised learning, foundation models for imaging, topological methods, or quantum machine learning.

Potential to transition into a longer-term or equity-based role as the company grows