This position will be based in the Department of Biomedical Informatics at Harvard medical school and will supervised by Andrew Beam, PhD and department chair Isaac Kohane, MD, PhD. The candidate will conduct original research on an exciting problem in the domain of medical question answering and will join a team of researchers working on similar problems in the area of natural language understanding. For this project we have curated a unique data resource of nearly 20,000 medical questions and over 200GB of medical text documents. The goal of this project is to create a cutting-edge system that is capable of answering medical questions posed as free text.
Candidates must have a PhD or other advance degree in statistics, computer science, biomedical informatics, or a related field. The candidate must have expertise in machine learning and experience programming deep learning models using either tensorflow/keras or pytorch.
Additionally, candidates are required to have expertise in deep learning and natural language processing as evidenced by at least one of the following:
- First-author publication in journal or conference in the field of deep learning or NLP
- First-author preprint of a deep learning/NLP project
- A well-documented github repository of a project written either in tensorflow or pytorch implementing a state of the art NLP model (such as a language model, transformer network, fast text, etc.) with an accompanying blog post demonstrating the concepts.
Strong preference will be given to candidates with previous experience in the area of automatic question answering. Experience with reinforcement learning, knowledge bases, and medical ontologies is also desired.
The candidate will be responsible for developing new methodology in the field of question answering and reducing these ideas to practice in python code. They will be expected to collaborate with other members of the lab, present their progress regularly at lab meetings, and to communicate their results to the scientific community through profession meetings and scientific publications.
As soon as possible
How to Apply
To be considered for this position, please email a single PDF document containing a letter of application that addresses your interest in and qualifications for the position, and a curriculum vitae. Please include the names and contact information of three references, who will be asked to supply letters or will be contacted by phone early in the application process. This is a one-year term appointment with the option to renew for a second-year conditional on satisfactory performance. The position is available now at Harvard Medical School in Boston.
Address Applications to:
Andrew Beam, Ph.D.
Instructor in Biomedical Informatics
Harvard University is an equal opportunity employer.
Women and minorities are especially encouraged to apply.