Internship - WSI/IO Inference Optimization

Want to help invent the future? Barco Labs is the innovation engine of Barco, where new technologies, concepts, and ideas are explored long before they become products.

About The Internship You’ll work on cutting-edge digital pathology solutions and help improve the performance of AI applications processing gigapixel medical images.

During this internship, you will investigate how large Whole Slide Images (WSIs) can be processed more efficiently for GPU-based AI inference. You'll analyse existing workflows, identify performance bottlenecks, and explore techniques to accelerate data transfer between storage and GPU memory. Your work may involve profiling, caching strategies, memory optimization, batching, prefetching, and leveraging GPU capabilities to improve end-to-end system performance.

You will collaborate with experienced software and AI engineers while contributing to real-world healthcare applications where performance and scalability are key.

What We're Looking For

• Bachelor or Master student in Computer Science, Software Engineering, Artificial Intelligence, Electronics-ICT, or a related field
• Strong programming skills in C++ and Python
• Interest in AI systems, performance optimization, and high-performance computing
• Knowledge of data structures, algorithms, and software development best practices
• Strong analytical and problem-solving skills
• Experience with Linux/Docker
• Experience with Git (GitHub) and collaborative software development
• Experience with LLM in development
• Fluent spoken and written communication in English

Experience with GPUs, Linux, Docker, Git, profiling tools, AI frameworks, or performance analysis is a plus.

What We Offer At Barco, you'll gain hands-on experience with modern AI infrastructure, large-scale medical imaging data, and advanced software optimization techniques. You'll work alongside experts in digital pathology, helping to build faster and more scalable AI solutions that support healthcare innovation.

Join us and help push the boundaries of AI performance in digital pathology.