Research Software Engineer – Clinical NLP (Data Science & AI Institute)

AIToolboard; Baltimore (MD); On-site; USD 120,000 - 180,000; Full time

Job summary

Johns Hopkins University is seeking a Research Software Engineer – Clinical NLP to design, implement, evaluate, and deploy advanced NLP pipelines for clinical text. The role involves working with clinicians, data scientists, and engineers to translate research into scalable software systems in secure environments.

Strong academic background and industry experience in NLP/LLMs, with a focus on privacy, robustness, and interpretability, are highly valued.

Qualifications

• Strong NLP, LLM, ML and DL skills.
• Experience building NLP models and pipelines in secure healthcare environments.
• Expertise in evaluating and benchmarking large language models.
• Ability to translate research into robust software systems.

Responsibilities

• Participate in ground-breaking research projects requiring advanced software solutions.
• Develop NLP pipelines for clinical text using modern DL libraries.
• Collaborate with clinicians and data scientists to meet project goals.
• Contribute to reproducible scientific workflows and documentation.

Skills

NLP

LLMs & Fine-Tuning

Prompt engineering

Software engineering

Healthcare NLP & HIPAA

Education

PhD/MS in CS/ML/Clinical NLP

Tools

PyTorch

TensorFlow

Job description

Jobs / Research Software Engineer – Clinical NLP (Data Science & AI Institute)

Research Software Engineer – Clinical NLP (Data Science & AI Institute)

Full-time

About the Role

Johns Hopkins, founded in 1876, is America's first researchuniversity and home to nine world-class academic divisions workingtogether as one university.The Johns Hopkins Data Science and AI Institute (DSAI) is a newpan-institutional initiative at Johns Hopkins to advance artificialintelligence and its applications, in part through investments inthe software engineering, data science, and machine learning space.DSAI is focused on revolutionizing discovery by advancingartificial intelligence that evolves collaboratively with humanintelligence, combining the strengths of each for the betterment ofsociety and the world in which we live. DSAI will bring togetherthe mathematical, computational, and ethical foundations of AI withthe domains of Health & Medicine, Scientific Discovery,Engineered Systems, Security & Safety, and People, Policy &Governance.DSAI seeks a Research Software Engineer - Clinical NLPSpecialty with strong academic background and relevant experiencein industry or academia focused on designing and buildingstate-of-the art clinical NLP systems. This position supportsresearch initiatives in the development and novel application ofNLP and large language models to extract insights from unstructuredclinical text using techniques such as named entity recognition(NER), negation detection, structured data extraction, diagnosisprediction, risk stratification, temporal reasoning andphenotyping. The successful candidate will play a critical role indesigning, implementing, rigorously evaluating, deploying andmaintaining robust and scalable NLP pipelines and models to extractmeaningful information from unstructured clinical text in secureenvironments, with the goal of enabling high-impact solutionsacross a range of biomedical domains. Experience with largelanguage models - such as fine-tuning, prompt engineering, modelevaluation, and adapting foundation models for domain-specificclinical tasks - is desirable, particularly in contexts that demandprivacy, robustness, and interpretability. The clinical NLP RSEwill work closely with clinicians, informatics researchers, datascientists and other RSEs to ensure NLP systems meet applicationgoals with methodological rigor and scientific reproducibility.DSAI engineers are at the forefront of modern data intensivescience, where professionally developed software is rapidlybecoming a key ingredient for success. The DSAI initiative includesthe build-out of a substantive and professional-scale softwareengineering capability, and a dramatic increase in infrastructure,both in hardware and in personnel. JHU has long been a world leaderin the broader domains of medicine and public health as well as awide range of science and engineering fields. This combined withour ethos of building out capabilities to have demonstrable globalimpact (e.g., JHUs Coronavirus Resource Center the award-winningglobal resource for real-time data and analysis for COVID-19) andother unique large scientific data sets, like the archives for theSloan Digital Sky Survey and several simulations, will be keyleverage points that will make the DSAI successful.Specific Duties & Responsibilities 0The successful candidates will participate in ground-breakingresearch projects that need advanced software solutions requiringexpertise in software engineering not commonly found in scientificcollaborations. 0The projects will require development of state-of-the artclinical NLP solutions using the latest deep learning librariestrained on state-of-the-art hardware in secure healthcare computingenvironments. 0Projects will involve analysis of massive data sets either inthe cloud or on premises. 0Projects will require development of novel NLP softwarepipelines for processing of unstructured clinical notes. 0Some projects may require deep engagement, possibly leading toco-authorship on scientific publications, while others may involvea more casual consulting engagement. 0They may require software solutions developed from scratch orrefactoring existing solutions to make them conform to industrystandards (quality, efficiency, reusability, robustness,portability, documentation, etc.). 0It is a high-level goal of DSAI to translate the efforts forthe individual projects into frameworks and template patterns forsustainable scientific infrastructure benefiting futureprojects.Special knowledge, skills, and abilities Strong NLP, LLM, machine learning and deep learningskills. Practical experience building NLP models and pipelines in asecure, HIPPA compliant healthcare environment. Expert-level knowledge of multiple modern NLP and LLM librariesand models. Hands-on experience adapting and fine-tuning large languagemodels for domain-specific clinical applications, with attention todata efficiency, interpretability, and reproducibility. Demonstrated expertise in prompt engineering, evaluation, andbenchmarking of large language models, includ