Research Assistant IV Non-Lab

at  Harvard University

JOB SUMMARY

The Department of Epidemiology, at the Harvard T.H. Chan School of Public Health, studies the frequency, distribution, and determinants of disease in humans, a fundamental science of public health. In addition to pursuing ground-breaking global research initiatives, we educate and prepare future medical leaders and practitioners as part of our mission to ignite positive changes in the quality of health across the world.
Dr. Andrew Beam’s research lab is principally concerned with improving, stream-lining, and automating decision-making in healthcare through the use of quantitative, data-driven methods. We do this through rigorous methodological research coupled with deep partnerships with physicians and other members of the healthcare workforce. As part of this vision, we work to see our ideas translated into decision-making tools that doctors can use to better care for their patients.
For more information on Dr. Beam’s lab and his research, please visit: http://beamlab.org/
The development of Cria, an artificial intelligence system specialized for neonatal medicine, has the potential to significantly advance health outcomes for vulnerable newborns. By creating a dedicated tool to navigate the exponentially growing literature, synthesize the latest evidence, and generate insights tailored to the neonatal field, this project promises to enhance clinical decision-making and accelerate the translation of discoveries into improved patient care.
In the short term, Cria will help address the challenge of information overload facing neonatal providers through efficient sorting and summarization of new research findings. This could allow physicians to more readily incorporate recent pivotal trial results and evolving best practices into their medical decisions and protocols. Over the longer-term, wide adoption of Cria across neonatal care units may help reduce variability in practice and promote greater adherence to evidence-based standards - impacting mortality, complications, and costs.
Additionally, this research will establish a general framework to adapt AI to other medical disciplines. The methods developed to transform an open-source language model into a reliable specialty-specific assistant can serve as a template for customizing tools across areas like cardiology, orthopedics, and more. By contributing an adaptable model and accessible tools, this project opens the door to powering precision medicine through AI.

POSITION DESCRIPTION

Reporting to Dr. Beam, the Research Assistant (RA) will support the development of Cria, an open-source AI system designed for neonatology, by contributing to machine learning and data science research efforts.

Duties and responsibilities include, but are not limited to, the following:

• Utilize deep learning frameworks such as PyTorch and TensorFlow, as well as Python and R programming languages, to perform data analysis and software development relevant to the Cria project.
• Create data visualizations and effectively communicate results to the research group and clinical collaborators, focusing on the application of large language models (LLMs) in neonatal medicine.
• Contribute to the development and refinement of machine learning models, including deep learning models for computer vision and natural language processing, to enhance Cria’s capabilities in neonatal medicine.
• Perform data cleaning, preprocessing, and exploratory data analysis in R and Python, with a focus on neonatal-specific datasets.
• Summarize and prepare results for manuscripts and research papers, contributing to the dissemination of findings related to the Cria project.

Other related duties may include assisting in the development of a general framework for adapting LLMs to specific medical domains and promoting open science and transparency in AI research.
PLEASE NOTE: This position has a term appointment of 8/30/2024. This is a part-time, benefits eligible position with a 20 hour work week.

BASIC QUALIFICATIONS

• 4+ years of related experience required; a combination of education and practical experience may be considered
• Experience working with pytho programming and the pytorch framework required

ADDITIONAL QUALIFICATIONS AND SKILLS

The following job-specific skills and competencies are preferred:

• Bachelor’s Degree is strongly preferred
• Master’s level training in epidemiology, biostatistics, public health, or related fields is a strong plus
• Excellent written and verbal communication
• Strong attention to detail, excellent time management and organizational skills, and experience prioritizing multiple tasks
• Experience working independently with minimal supervision and working collobratively with a team

The following cultural competencies are also preferred:

• Awareness of and aptitude to appropriately and effectively understand, respect, and adapt to cultural and identity-based difference within group environments
• Knowledge of and commitment to concepts and issues tied to social justice, diversity, equity, and inclusion
• Skills related to creating and supporting an environment that allows for inclusion, effective intercultural engagement, and personal humility and authenticity
• Experience fostering and reinforcing an environment that values unique experiences, cultures, backgrounds, and goals

• Utilize deep learning frameworks such as PyTorch and TensorFlow, as well as Python and R programming languages, to perform data analysis and software development relevant to the Cria project.
• Create data visualizations and effectively communicate results to the research group and clinical collaborators, focusing on the application of large language models (LLMs) in neonatal medicine.
• Contribute to the development and refinement of machine learning models, including deep learning models for computer vision and natural language processing, to enhance Cria’s capabilities in neonatal medicine.
• Perform data cleaning, preprocessing, and exploratory data analysis in R and Python, with a focus on neonatal-specific datasets.
• Summarize and prepare results for manuscripts and research papers, contributing to the dissemination of findings related to the Cria project