The Allen Institute for AI
Young Investigator, Open Language Models for Biology
$160K
Seattle, WA +1 · Full-time · Posted Jul 28, 2026
- Location
- Seattle, WA
Seattle, WA, United States - Experience
- No specific number of years stated; PhD completed or expected within one year.
- Education
- PhD in machine learning, computational biology, computational neuroscience, physics, or a related quantitative field.
- Visa sponsorship
- The posting welcomes applicants from outside the United States and states that the employer participates in E-Verify; it does not specify a visa sponsorship policy.
About the job
Develop and evaluate open multimodal language models that combine single-cell brain data with biological text to investigate questions about Alzheimer’s disease and neurodegeneration.
What you'll do
- Build, adapt, and evaluate models for single-cell, brain-region, and donor-level data, including controlled tests of language grounding.
- Collaborate with Allen Institute scientists to interpret results and identify predictions for experimental follow-up.
- Release research data, code, and model weights, and prepare findings for publication.
What they ask for
- Completed PhD, or PhD expected within one year, in a relevant quantitative field.
- Hands-on experience training or fine-tuning transformer-based language models.
- Strong Python and PyTorch skills, plus experience writing reliable, reproducible research code.
- Ability to conduct independent research and communicate clearly across machine-learning and biological disciplines.
Nice to have
- Experience with single-cell genomics, tools such as scanpy, AnnData, or the scVI ecosystem, or neurodegeneration and brain atlases.
- Experience with multimodal models, language-model post-training, or public research releases.
Pay
159,650.00 USD base salary
Tools
Python, PyTorch, OLMo, scGPT, Geneformer, CellWhisperer, C2S-Scale, scanpy
AI focus
Develop open multimodal language models combining single-cell gene-expression data with text; test whether biological language grounding improves performance and evaluate representations of brain regions, donors, pathology, and disease progression.
Model-written summary · the original posting is authoritative