Founding Research Scientist, Biological Foundation Models
Hybrid (New York City, NY)$150k - $250k + equityFull-timePosted 1mo ago
About the role
Founding role responsible for end-to-end modeling in CellType's biology programs, focusing on biological foundation models and agent systems. Hands-on with data, modeling, evaluation, and collaboration with partners to translate scientific questions into deployable model capabilities.
Responsibilities
- Own modeling for one or more live pharma and biotech programs: framing the problem, choosing the representation and architecture, building the training and evaluation loop, and delivering results to the partner
- Train, fine-tune, and post-train biological foundation models on single-cell, perturbation, spatial, multi-omic, imaging, and clinical assay data
- Build models of cellular state and response: perturbation prediction, in-silico screening, cross-species and cell-line-to-patient translation, deconvolution, and related virtual-cell tasks
- Design evaluations that biologists trust: held-out perturbations, unseen cell types and tissues, cross-species and cross-cohort generalization, and wet-lab-in-the-loop validation with our partners
- Decide what data to acquire or generate next, and shape the data-generation experiments we run with academic and pharma collaborators
- Turn one-off project results into reusable model capabilities and product workflows
- Present results to partner scientists and translational teams, and defend the modeling choices behind them
- Work alongside our model-training and platform engineers so that research ideas become reliable, scalable systems
Qualifications
- Have trained or materially improved large models yourself, in any domain: vision, multimodal, language, speech, protein/molecule, or biology
- Strong applied ML judgment: representation choices, encoding of non-text data, loss design, pretraining vs. fine-tuning vs. post-training
- Fluent in Python and PyTorch or JAX, comfortable on multi-GPU infrastructure
- Have shipped models that people relied on in production or in a live program
- Worked hands-on with biological data such as scRNA-seq, Perturb-seq / CRISPR or compound screens, spatial transcriptomics, bulk or cell-free omics, histopathology, or high-content imaging
- Understand biology behind the data to assess plausibility and choose sensible baselines
- Followed or contributed to virtual-cell / perturbation-modeling literature and have opinions on what works
Skills
Required:PythonPyTorchBiology knowledge (single-cell, Perturb-seq/CRISPR, spatial omics)
Last updated Oct 22, 2026. Interested? Apply on the company site.
About CellType
The agentic drug company. We simulate human biology.
CellType is building an agentic drug company. AI agents that run the full drug discovery pipeline on top of biological foundation models that simulate human biology. Our core technology, developed with Google DeepMind, has already discovered and validated a new cancer treatment signal. We’re working with Top 10 pharma.