Joel Dapello

Joel Dapello is an engineer and scientist who has worked at the intersection of Bio, Neuro, and AI for over ten years, developing methods to accelerate biological sciences. During his career, Joel has worked on a wide range of problems, from developing novel wet-lab methods to analyzing and modeling high-dimensional biological data, and from training biological foundation models to developing agentic therapeutic target assessment platforms. Joel prioritizes impact, follow through, and clear communication.

Joel was the founding engineer at BioBright (acquired by Dotmatics), where he worked with Charles Fracchia and Adam Marblestone to develop a lab-integrated electronic lab notebook and the first end-to-end encrypted data collection and analysis platform for life sciences. He completed his PhD in Applied Mathematics at Harvard University working with Jim DiCarlo and David Cox studying the high-dimensional geometry of real and artificial neural representations and developing brain-inspired methods to improve adversarial robustness in deep learning models.

Currently, Joel is a machine learning scientist and technical lead at Altos Labs, where he worked with Thore Graepel and Morgan Levine to found and develop the multimodality foundation model program. He now leads a cross-functional team of biologists and ML engineers developing and deploying agentic AI systems for therapeutic target assessment and prioritization.