Research Scientist (AI)\_Cell & Tissue Modeling
Palo Alto, Paris, Abu Dhabi Engineering / Full Time / On-site
Key Responsibilities
+ PhD (or evidence of equivalent level of expertise) in Computer Science, Artificial Intelligence, Machine Learning, or a related technical field.
+ Proven track record in research and innovation demonstrated through contributions in top-tier AI/ML (e.g., NeurIPS, ICML, CVPR, ECCV, ICCV, ICLR) and/or core biology (e.g., Nature, Science, or Cell) journals and conferences.
+ Skilled in developing, implementing, and debugging deep learning methods/models in popular frameworks, such as JAX, TensorFlow, or PyTorch, with an interest in generative models, graph neural networks, or large-scale deep learning applications.
+ A strong theoretical foundation (statistics, optimization, graph algorithms, linear algebra) with experience building models ground up.
+ A passion for interdisciplinary research (with an emphasis on the intersection of AI and Biology), and willingness to acquire necessary domain knowledge.
+ Motivated and self-driven with the ability to operate with partial and incomplete descriptions of high-level objectives (as is typical in a start-up environment).
+ Evidence of familiarity and utilization of software engineering best practices (version controlling, documentation, etc), and open-source contributions, especially if used by others.
+ 3+ years of post-PhD experience in an industry or postdoc role
+ Prior experience working at either a start-up or top research industry labs (e.g., OpenAI, FAIR, Deepmind, Google Research).
+ Hands-on prior experience working at the intersection of AI and Biology.
+ Experience in large-scale distributed training and inference, ML on accelerators.
Preferred Qualifications
+ Experience with cell-level data, particularly single-cell RNA-sequencing data.
+ Experience with tissue-level data, particularly spatial transcriptomics, spatial proteomics, or microscopy (e.g. H&E, IF, IHC).
+ Experience with methods development for afore-mentioned data types.
+ Experience with multimodal or multiscale models (even in other domains, e.g. remote sensing, medical imaging).
+ Deep knowledge of one or more of the following: variational autoencoders (especially biological variants like scVI), vision transformers, graph neural networks, neural fields, diffusion models, and self-supervised learning.
