Scientist - ML​/AI algorithms Multimodal Foundational Models

Scientist I – ML/AI Algorithms for Multimodal Foundational Models

Join to apply for the Scientist I – ML/AI Algorithms for Multimodal Foundational Models role at Allen Institute
.

Scientist I – ML/AI Algorithms for Multimodal Foundational Models – Human Cell Types

The mission of the Allen Institute is to understand the principles that govern life, and to advance health. Our creative and multi‑dimensional teams focus on answering some of the biggest questions in bioscience. We accelerate foundational research, catalyze bold ideas, develop tools and models, and openly share our science to make a broad, transformational impact on the world.

The mission of the Allen Institute for Brain Science is to accelerate the understanding of how the human brain works in health and disease. Using a big science approach, we generate useful public resources, drive technological and analytical advances, and discover fundamental brain properties through the integration of experiments, modeling, and theory.

The Seattle Alzheimer’s Disease Brain Cell Atlas (SEA-AD) consortium strives to achieve a deep molecular and cellular understanding of the early pathogenesis of Alzheimer’s disease (AD). Leveraging advances in quantitative neuropathology and next‑generation single‑cell molecular profiling technologies, we are generating large‑scale data sets from well‑characterized donors that span the spectrum of AD pathology to comprehensively map human brain cell types across aging and disease.

These data along with tools for their use are made freely and publicly available.

We seek to hire a Research Scientist to design modern machine learning methods to integrate multimodal data, infer gene regulatory mechanisms, describe omics trajectories and contribute to the mechanistic understanding of Alzheimer’s disease pathology. The successful candidate will have a strong background in computational biology, computer science, or statistics, and experience developing and training machine learning models (deep generative models, score‑based models, transformer architectures).

Strong preference will be given to individuals with a track record of both individual and team contributions in solving complex research problems, and experience in cutting‑edge computational methodologies applied to biological –omics, spatial, and/or clinical metadata.

• Develop modern machine learning algorithms to model multimodal omics data (transcriptional, chromatin accessibility, genetic, spatial, and protein information)
• Evaluate ML architectures able to characterize regulators of cell identity and neurodegenerative progression
• Implement sequence‑based models to predict genetic contributions to disease
• Train large‑scale models in multi‑GPU environments efficiently
• Stay at the forefront of advances in AI for multimodal molecular biology
• Participate in a highly interactive and multidisciplinary environment
• Publish/present findings in peer‑reviewed journals/scientific conferences

• Ph.D. in Computer Science, Applied Math, Engineering, Computational Neuroscience, Computational Biology, or related field, or equivalent combination of degree and experience
• Experience working with recent Deep Learning Architectures/Foundational Models
• Experience with building, deploying and maintaining machine learning and foundational models
• Proficiency with cloud computing (AWS, GCP or Azure) and with on‑prem clusters
• Proven ability to work independently and manage multiple projects simultaneously while meeting deadlines

• Experience with modeling gene regulatory networks, chromatin accessibility information, or related computational biology studies
• Experience building transformer‑based foundational models of cell identity
• Working knowledge of AI/ML custom libraries, AI/ML execution platforms
• Strong publication track record
• Proven ability to work independently and manage multiple projects simultaneously while working in a highly collaborative environment
• Excellent written and verbal communication skills, with the ability to collaborate effectively in a multidisciplinary team environment

• Occasional…