Teams at Biohub

Build infrastructure for biological foundation models and large-scale machine learning. Responsibilities include training and evaluating models, curating petabyte-scale datasets, developing shared libraries, optimizing GPU data loading, maintaining distributed training systems, deploying and monitoring models, tracking artifacts, and improving testing and continuous integration.

Design and build petabyte-scale data pipelines for genomic and imaging data, transforming public and proprietary datasets into AI-ready training data. Develop agent-based curation and quality-control systems, data cataloging tools, and reliable observable infrastructure. Collaborate with AI researchers and scientists to define data requirements and integrate diverse biological datasets while addressing performance, bandwidth, scalability, and scientific correctness.

Build and scale frontier AI systems for biology: design model architectures, run large-scale pretraining on GPU clusters, develop RL/reward modeling and multi-agent systems, create evaluation frameworks tied to biological outcomes, and publish/open-source research.