Stephen Tashobya is an accomplished AI Engineer with deep expertise in machine learning, computer vision, and biomedical signal analysis, specializing in digital health and physiological signal modeling. His work bridges artificial intelligence and clinical practice, delivering interpretable, scalable, and clinically impactful solutions for cancer diagnosis, maternal-fetal health, and precision medicine.

He has extensive experience designing and deploying end-to-end deep learning pipelines for whole-slide image (WSI) classification, tumor segmentation, and histological patch analysis. Stephen has successfully implemented advanced architectures such as ResNet, DenseNet, U-Net, HoVer-Net, and Vision Transformers, and has applied foundation models including CLIP and SAM in medical imaging contexts.

Beyond imaging, he is highly skilled in time-series modeling using recurrent neural networks (RNNs) for physiological signal analysis, including electrohysterograms (EHG) and tocograms for predicting preterm birth risk. He leverages Generative AI techniques such as diffusion models and GANs for synthetic data generation and data augmentation, particularly in low-resource medical domains.

Professional Experience

AI Engineer – Inventlab, Cleveland, USA

Stephen develops high-performance AI pipelines for digital pathology, including patch extraction, tumor classification, and heatmap generation. He collaborates with pathologists to enhance model interpretability and integrates multi-GPU inference systems for large-scale datasets.

Founder & CEO – Wekebere Company Ltd

Stephen leads a multidisciplinary health-tech company focused on reducing maternal and neonatal mortality through noninvasive AI-driven monitoring solutions. Under his leadership, the company has secured international funding, built strategic global partnerships, and developed wearable maternal health technologies powered by machine learning.

Education

• MSc Biomedical Engineering – Case Western Reserve University (2025)

• MSc Data Communication & Software Engineering – Makerere University (2022)

• Certificate in Health Innovation & Entrepreneurship – University of Oxford (2021)

• Certificate in Evidence-Based Medicine & Public Health – Stellenbosch University (2018)

• BSc Software Engineering – Makerere University (2017)

Technical Expertise

• Programming: Python, C++, MATLAB, R

• Deep Learning: PyTorch, TensorFlow, Keras

• Medical Imaging: MONAI, OpenSlide, SimpleITK, OpenCV

• Cloud & Compute: AWS EC2, GCP, SLURM, CUDA, multi-GPU training

• Deployment: Docker, Flask, ONNX, TorchScript

• Data Handling: NumPy, Pandas, HDF5, Zarr, DICOM

Awards & Recognition

Stephen is a multi-award-winning innovator recognized internationally for his contributions to digital health and AI innovation, including:

• Winner – Johns Hopkins Healthcare Design Competition

• Winner – Africa App Launchpad

• 1st Runner-Up – Digital Equality Award

• Winner – Youth Connect Africa Technology & Inclusivity Award

• RUFORUM Top 15 Young African Entrepreneurs Award

• Representative of Uganda innovators at World Mobile Congress, Barcelona