Staff Scientist 1 – Literature Development Team

at  National Library of Medicine

The National Library of Medicine (NLM) is one of 27 Institutes and Centers at the National Institutes of Health. NLM is a global leader in biomedical informatics and computational health data science and the world’s largest biomedical library. NLM’s legislative mandate is to support the essential work of acquiring, organizing, preserving, and disseminating biomedical information. NLM provides public access to this information 24 hours a day, seven days a week.
NLM Division: National Center for Biotechnology Information (NCBI)
NLM Program/Office: Information Engineering Branch (IEB)/Literature Development Team
Application Period: Applications will be accepted August 6, 2024 through September 6, 2024.
Overview:
The IEB is responsible for designing and building NCBI’s production software and databases. They: (1) perform applied research in the area of data representation and analysis for molecular biology data, including development of computer based systems for the storage, management, and retrieval of knowledge concerning molecular biology, genetics, and biochemistry; (2) design schema and specifications for representation of sequence genetic and structural information in databases which will serve as national resources; (3) design and develop distributed software systems from the prototype to operational phases which provide researchers with both local and remote computational services; (4) coordinate access to sequence, genetic, structural, and bibliographic databases; (5) establish collaborative informatics research projects with NIH Intramural laboratories as well as extramural academic groups; (6) consult and advise other government agencies and research laboratories on advanced methods of software and database design; and (7) develop and promote standards for databases, data exchange, and biological nomenclature.
NCBI programs and activities supported by this position are highly complex, mission-critical, and have national and international impact, as they directly support dissemination of biomedical information in service to NIH and the broader biomedical community through NCBI, an organization with a budget of more than $150 million and a staff of more than 700 Federal and contract staff.
Position Description/Responsibilities:
The position of Staff Scientist 1 is located in IEB’s Literature Development Team. This team is responsible for all Literature related databases and programs at the NCBI. These include PubMed, an index of life sciences journal literature of over 35 million records, PubMed Central (PMC), a free full-text digital archive of life sciences journal articles, and the NCBI Bookshelf, a repository for non-journal literature. Additionally, the Literature Team supports the NIH Public Access policy and the NIH Manuscript Submission system, which prepares manuscripts resulting from NIH-funded research for inclusion in PMC. This program integrates the deposited literature with other NCBI resources, such as GenBank, and provides facilities for accessing the deposited literature and links to related resources.
The Senior Machine Learning Computer Scientist (software engineer) will be responsible for the continued successful development and maintenance of the Medical Text Indexer (MTI) system. The selectee will focus on designing and implementing sophisticated machine learning models to further improve the quality and usefulness of automatic Medical Subject Headings (MeSH) indexing in PubMed.
MTI is the main product of the Indexing Initiative project and has been providing indexing recommendations based on the MeSH vocabulary since 2002. In 2011, NLM expanded MTI’s role by designating it as the first-line indexer (MTIFL) for a few journals; today the MTIFL workflow includes over 350 journals and continues to increase. The close collaboration of the NLM Index Section, Lister Hill National Center for Biomedical Communications, and Office of Computer & Communications Systems continues to expand and refine the ability of MTI to provide assistance to the indexers.
The ideal applicant must be able to use new technologies and methodologies to address the difficulties associated with medical text indexing and will possess the background and abilities needed to stay current with the rapidly evolving fields of artificial intelligence and machine learning. He/she/they will be essential in maintaining the correctness and efficiency of the current system by optimizing it and providing continuous assistance.
The selectee will have the opportunity to participate in trans-NLM and/or trans-NIH projects and committees, serve in outreach-associated or staff-training activities and lead workgroups or teams, such as those that design or influence improvements in program policies, processes, or other key activities. Ideal candidates will work with a diverse group of scientists, bioinformaticians and other developers across the library to maintain and continually improve the NLM’s new machine learning based Medical Text Indexer system (MTIX). They will leverage cloud-based architectures and technologies to deliver optimized machine learning models at scale.

POSITION REQUIREMENTS:

The ideal candidate may or may not be a United States citizen and must have a doctoral degree.

• Ph.D. in Computer Science, Engineering, Physics, or Applied Mathematics.
• 15+ years of relevant computer programming experience in a Windows or Linux environment.
• 6+ years of on-the-job experience with an industry recognized machine learning framework (e.g., PyTorch or TensorFlow).
• Proficiency in the Python programming language.
• Expert in deep learning for natural language processing.
• Knowledge and experience of Large Language Models (LLMs) for natural language processing (e.g., BERT).
• Proven track record developing machine learning systems to solve real-world problems.
• Experience of the full machine learning project lifecycle including dataset creation, model training, model evaluation, and model deployment.
• Experience of neural network training and inference using high-performance computing resources (e.g., NIH BioWulf).
• Experience deploying machine learning models to cloud computing environments (e.g., using Amazon Sagemaker).
• Knowledge of controlled vocabularies (e.g., MeSH) and their application to document indexing (e.g., MEDLINE indexing).

EDUCATION REQUIREMENTS:

Selectees who have completed part or all their education outside of the United States must have their foreign education evaluated by an accredited organization to ensure that the foreign education is equivalent to education received in accredited educational institutions in the United States. We will only accept the completed foreign education evaluation. For more information on foreign education verification, visit the National Association of Credential Evaluation Services (NACES) website. Verification must be received prior to the effective date of the appointment.
Salary and Benefits:
Salary is commensurate with research experience and accomplishments. A full package of benefits, including retirement, health, life, and long-term care insurance, Thrift Savings Plan participation, etc., is available.
The successful candidate will serve in a non-competitive, time-limited, renewable appointment in the excepted service. Review our benefits

Please refer the Job description for details