Arsen Abzhanov

I am a computer vision researcher and AI engineer with a strong academic background and hands-on experience building explainable medical diagnosis systems and production ML tools. I hold an MSc in Computer Vision (expected 2025) and completed a BSc in Computer Science, backed by a full scholarship and focused research in Explainable AI for medical diagnosis.

I have delivered applied AI projects and intern roles that span model research, multimodal vision-language work, and cloud/DevOps deployments — improving diagnostic model metrics, implementing RAG chatbots, and building automated data-processing pipelines for financial services. Notable achievements include boosting classification accuracy and dice scores in a MICCAI challenge (3rd place), integrating a diffusion transformer to improve robustness and F1 across skin-lesion and ultrasound datasets, and publishing at a MICCAI workshop.

I work primarily with Python, PyTorch, transformers and modern MLOps tooling (Docker, Azure, AWS, LangChain), and I prioritize explainability, data security, and production-ready deployments. I seek roles where I can advance clinical AI solutions and help operationalize research into reliable products.