Associate Computational Scientist - AI ML

📋 Description

• • Join the AI/ML group inside JAX Data Science and become the bridge between cutting-edge machine-learning research and life-changing biomedical discovery. You will wield state-of-the-art AI to decode imaging and multi-omics data that span cancer, aging, longevity, and neurodegeneration—turning petabytes of raw information into insights that accelerate cures.
• • Design and execute end-to-end data pipelines that ingest, harmonize, and QC single-cell RNA-seq, spatial transcriptomics, ATAC-seq, proteomics, and high-resolution microscopy or MRI volumes. You will curate metadata, enforce FAIR principles, and build reproducible workflows that empower bench scientists to explore data without bottlenecks.
• • Prototype and refine deep-learning models—CNNs for histopathology, transformers for multi-modal integration, and generative architectures for data augmentation—using TensorFlow/Keras or PyTorch on JAX’s GPU clusters. You will benchmark against gold-standard datasets, tune hyper-parameters with Bayesian optimization, and package models as containerized micro-services for internal and external users.
• • Pioneer spatial-omics analytics that map gene expression to tissue architecture across mouse and human samples. You will extend existing tools (e.g., Seurat, Scanpy, STUtility) and invent new algorithms that align brains to the Allen Atlas or tumors to 3-D organoid scaffolds, enabling cross-species translation of cellular phenotypes.
• • Serve as computational lead on 3–5 concurrent grant-funded projects, translating PI goals into sprint plans, JIRA tickets, and GitHub milestones. You will anticipate risks, allocate GPU hours, and deliver interim dashboards that keep multidisciplinary teams—from oncologists to mouse geneticists—aligned and engaged.
• • Author first-author and co-author manuscripts for high-impact journals, transforming exploratory notebooks into publication-ready figures with R Markdown or Quarto. You will also draft bioinformatics sections of NIH and foundation proposals, translating complex AI concepts into language that resonates with study sections.
• • Mentor junior analysts and rotating graduate students through code reviews, pair programming, and weekly “AI clinic” office hours. You will curate internal training materials—Jupyter-based workshops, Snakemake templates, and Docker recipes—that scale expertise across the institution.
• • Present findings at conferences (e.g., ISMB, AACR Genomics, MICCAI) and represent JAX in external consortia such as the Human Tumor Atlas Network and the Cellular Senescence Network. Your talks and posters will showcase JAX’s leadership in AI-driven biomedical discovery and attract new collaborations.
• • Contribute to JAX’s strategic initiative of integrating human and mouse biology through cellular models by ensuring that every algorithm you build is benchmarked in both species. Your work will directly inform JAX’s therapeutic pipeline, from validating CRISPR screens in murine systems to predicting drug responses in patient-derived xenografts.
• • Maintain rigorous software-engineering standards: unit tests with pytest, continuous integration via GitHub Actions, semantic versioning, and comprehensive documentation. You will champion open science by releasing non-sensitive code under MIT/GPL licenses and publishing interactive notebooks on JAX’s public GitHub portal.
• • Stay at the forefront of AI/ML innovation by attending internal reading groups, piloting emerging frameworks (e.g., JAX, Lightning, Dask-cuDF), and evaluating cloud-native solutions on AWS and Google Cloud. You will advise IT on infrastructure investments that keep JAX competitive in an era of exponentially growing data.