2027 Future Talent Program - Patient Data Learning - Co-op
About The Position
The Future Talent Program within the Data, AI, and Genome Sciences (DAGS) Department offers a six-month co-op opportunity in the Patient Data Learning group in Precision Genetics, based in Cambridge, Massachusetts. The computational project is intended to help students further develop their skills and evaluate whether the company is a fit for their long-term goals. The co-op will involve designing and building an agent-assisted system or components to support feasibility assessment and early exploration of patient phenotypes and clinical profiles in EHR-linked biobanks. The work includes reusing and customizing AI language models to translate natural-language scientific questions into machine-generated queries evaluated in agentic loops; creating guardrails and evaluations to promote and measure model and agent reliability; and applying these tools and technologies to answer feasibility questions.
Requirements
- Currently enrolled in at least a master's degree program in Clinical Informatics, Epidemiology, Computational Medicine, Biomedical Informatics, Biostatistics, Computer Science with a strong interest in biomedical applications, or a related field. Students enrolled in a PhD program are highly encouraged to apply.
- Computational and software engineering fundamentals sufficient to run and tune local LLMs, build local agentic systems, and apply or outline software and agent development practices such as version control, packaging, tests, and evaluations.
- Proficiency in Python and SQL; database-agnostic querying such as ibis is ideal.
- Applied knowledge sufficient to query structured EMR/EHR data, such as creating patient features, building cohorts, or reporting summary statistics, or demonstrated interest in learning to do so and in clinical applications.
- Availability to work full-time for six months in 2027–2028 as part of a formal co-op program.
- Excellent academic achievement and analytical ability.
- Superior communication and interpersonal skills, and ability to work well in a collaborative team environment.
Nice To Haves
- General interest or experience in clinical informatics, clinical machine learning, computational medicine, natural language processing and LLM research, or epidemiology.
- Good data systems fundamentals, with at least a basic understanding of data engineering and database-agnostic design.
- Knowledge of agent reward systems.
- Knowledge of statistical methods for epidemiology or econometrics.
- Familiarity with the OMOP CDM and medical coding systems such as ICD, SNOMED, CPT, and LOINC.
- Experience with Databricks.
- Knowledge of R and tidyverse packages.
Responsibilities
- Design and build an agent-assisted system or components for feasibility assessment and early exploration of phenotypes and clinical profiles of patients in EHR-linked biobanks.
- Reuse and customize AI language models to translate natural-language scientific questions into machine-generated queries evaluated in agentic loops.
- Create guardrails and evaluations to promote and measure the reliability of models and agents.
- Assist with applying these tools and technologies to answer feasibility questions.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Intern
