Summary

Design and build large-scale multimodal foundation models (Vision Transformers, Multimodal LLMs) that fuse high-dimensional biomedical imaging, molecular, and language data. Implement cross-domain data fusion strategies and scale ML pipelines to process petabyte-scale image repositories and multi-omics streams on distributed cloud infrastructure. Collaborate with experimental scientists and engineers to translate biological complexity into production-ready, performant code. Requires PhD in Computer Science, AI/ML, or related field; deep expertise in modern computer vision architectures, distributed training frameworks (PyTorch Distributed, DeepSpeed, Jax), and expert-level Python for production ML systems.