The Luna lab is jointly affiliated with the National Library of Medicine (NLM) and the National Cancer Institute (NCI). The dual ambitions of the lab are to make biomedical data and information accessible and to advance cancer research to help people live longer, healthier lives. We seek outstanding, highly motivated, and skilled candidates to join our team to explore how researchers interpret large data sets and enable them to use and gain insight from data sets through interactive systems. About the position This position offers a unique opportunity to push beyond the traditional scope of data extraction and management techniques to address the new challenges of large-scale, heterogeneous data. The group combines expertise in data management, visualization, information retrieval, and network data. The candidate will work in areas of data integration, data uncertainty, data summarization, to explore data query methods that leverage multimodal vision large language models (vLLMs). Through this work, selected applicants will contribute collaboratively to research in the field of cancer therapeutics and precision medicine that spans both basic and translational science objectives. Successful candidates will develop novel bioinformatic machine learning methodologies in the areas of: Multimodal data summarization from pharmacogenomics and clinical notes datasets, Structuring complex scientific information from unstructured manuscript data (e.g., supplementary data, experimental protocols, images) and Retrieval-augmented generation for LLM-backed agents for information gathering from cancer-related multi-omics datasets. We are offering full-time postdoctoral fellow positions, available immediately and renewable on a yearly basis. Initial appointments will be for 1 year, with possible extensions up to 5 years. The NIH offers a competitive stipend, a stipend supplement for comprehensive health insurance, and is dedicated to the continued education and career development of all its research staff. These positions are subject to background checks. Recent Articles CellMiner Cross-Database for

Pharmacogenomics BioFactoid:

Retrieving Pathway Knowledge Parsing Systems Biology Graphical Notation Maps Additional Links Augustin Luna - CCR staff profile , NLM profile , Google Scholar Summary of NIH as a Training Environment (Salary, Reputation, Cost of Living, Social Climate) About the National Institutes of Health (NIH) The National Institutes of Health is made up of 27 separate institutes and centers that include the National Library of Medicine (NLM) and the National Cancer Institute (NCI). About the NLM IRP The National Library of Medicine (NLM) pioneers new ways to make biomedical data and information more accessible; and builds tools for better data management and personal health. NLM's cutting-edge research and training programs (with a focus on artificial intelligence (AI), machine learning, computational biology, and biomedical informatics and health data standards) help catalyze basic biomedical science, data-driven discovery, and health care delivery. About the

NCI/CCR/DTB

The National Cancer Institute Center for Cancer Research (NCI-CCR) is the largest division of the NCI; it encompasses various branches such as the NCI Developmental Therapeutics Branch. The NCI CCR has a mandate to confront the special challenges presented by rare cancers as well as cancers that may be predominant in medically underserved populations. One way in which the NCI CCR addresses this mandate is by conducting clinical trials that recruit patients with rare cancers thereby generating unique data to advance research in these cancers. While rare cancers affect low numbers of patients, as a group, they account for about a quarter of all cancers, as well as a quarter of all cancer deaths each year . What you'll need to apply To , please send the following: Cover letter (1 page max) describing your 1) research experiences, 2) training goals, and 3) preferred starting date. Mention projects or articles of the Luna group of interest and explain your potential role Updated CV including bibliography It is suggested that links to a code repository URL(s) be included in your application with code attributable to the applicant Contact information (name, institute, email, phone) for 3 references to Augustin Luna, Ph.D., via email. Write "Postdoctoral Application" in the subject heading. If we are interested, you will be contacted by Dr. Luna. Contact name Augustin Luna Contact email augustin@nih.gov Qualifications Essential PhD in a relevant field, including: Bioinformatics, Biomedical Engineering, Mathematics, Data Science, Computer Science, Medical Informatics, or a degree related to Biology with substantial experience in computational and statistical work. Individuals in the final stages of PhD submission will be considered as well as PhD graduates within 5 years of graduation. Knowledge of theory and practice of LLM and foundation models, as well as deep learning neural networks Strong knowledge and experience in coding (R/Python or similar languages) Technical expertise in machine learning and/or mathematical modeling An interest in applying computational methods to biological problems A demonstrated ability to generate and pursue independent research ideas Excellent communication skills, written and verbal as evidenced by publications, preprints, and/or conference presentations in conversational artificial intelligence venues (e.g., CoLing, EMNLP, ACL, NAACL, IJCAI, ICLR, NeurIPS, AAAI, CVPR, IEEE, JAMIA, etc) Dedication to reproducible research and open science Desirable Foundational knowledge in Bioinformatics, Systems Biology, and/or similar fields Foundational knowledge in Mathematics, Statistics, and/or Data Science Familiarity with software development practices Experience in conversational interface design for real-time interaction Experience multimodal generative language models, personalized LLM, and/or fine-tuning LLMs with/for reinforcement learning planning Experience with analysis using the R programming language Experience with using network-based analyses (graph theory) and software/resources (graph and/or pathway databases) is highly desirable Experience with biomedical ontologies Development and execution of annotation tasks with teams of experts Experience working in collaborative interdisciplinary environments Disclaimer/Fine Print U.S. citizens and permanent residents are eligible to apply.