Open-Language Biology ML Postdoc (Single-Cell)

Ai2; Seattle (WA); On-site; USD 143,685 - 175,615; Full time

Benefits offered by this job

Medical, dental, vision coverage

401(k) plan

Commuting stipend and wellness stipend

Generous vacation and sick leave

Job summary

Ai2 in Seattle seeks a Postdoctoral Young Investigator to advance CellOLMo, an open science project combining single-cell biology with language models. You will design experiments, train models, and release code and data to the research community.

You should have a PhD in a quantitative field and hands-on experience with transformer models, Python, and PyTorch. Collaboration with the Allen Institute is a core part of the role, with access to substantial GPU compute.

Qualifications

• PhD in ML, computational biology, neuroscience, physics or related field.
• Hands‑on experience training or fine‑tuning transformer‑based language models.
• Strong Python and PyTorch skills.
• Experience writing reliable, reproducible research code.
• Ability to conduct independent research and communicate with scientists.
• Willingness to release code and data openly.

Responsibilities

• Develop a multimodal language model combining single-cell data with text.
• Reproduce and evaluate scGPT, Geneformer, CellWhisperer, and C2S-Scale.
• Adapt approaches for brain and neurodegeneration data.
• Design controlled experiments to test grounding improvements.
• Develop representations at brain-region and donor levels.
• Evaluate models on held-out donors with metrics like cell-type recovery and regional patterns.
• Collaborate with Allen Institute scientists to interpret results.
• Release data, code, and model weights; prepare results for publication.

Skills

Python

PyTorch

Transformer models

Independent research

Scientific communication

Education

PhD in ML/Computational biology/Neuroscience/Physics

Job description

Ai2 in Seattle seeks a Postdoctoral Young Investigator to advance CellOLMo, an open science project combining single-cell biology with language models. You will design experiments, train models, and release code and data to the research community.

You should have a PhD in a quantitative field and hands-on experience with transformer models, Python, and PyTorch. Collaboration with the Allen Institute is a core part of the role, with access to substantial GPU compute.