The Data Engineer will work closely with clinical researchers, biostatisticians, data scientists, and cross-functional teams to translate complex research and clinical questions into scalable data solutions. In addition, this role will collaborate side by side with data engineers across the broader Montefiore data organization, aligning with enterprise data standards, platforms, and best practices to ensure consistency, scalability, and interoperability across clinical, operational, and research domains.
This role also contributes to the advancement of enterprise research data platforms, supporting secure data access, governance, and innovation in analytics to improve patient outcomes and accelerate scientific discovery.
Responsibilities
Develop and maintain research data pipelines that integrate clinical records, text, imaging, and other approved data sources. Document data definitions, quality checks, metadata, lineage, and transformations to support FAIR aligned reuse.
Help implement standardized, governed data access workflows and reusable tools for cohort discovery, phenotyping, trial feasibility, recruitment planning, and research analytics. Translate investigator needs into scoped requirements and track delivery.
Develop, benchmark, and validate AI methods, including information extraction, multimodal models, large or vision language models, and interpretable or knowledge informed approaches when suitable for the use case.
Turn validated methods into maintainable research services and interfaces, with version control, automated tests, monitoring, user documentation, and clear handoff to authorized operational partners.
Set evaluation plans with investigators and clinical experts, including baselines, validation splits, human review, error analysis, reproducibility checks, and assessment of subgroup performance and potential bias.
Work with ICTR, the Data Science Institute, and Montefiore data, IT, security, privacy, and oversight partners to implement approved access and governance requirements. Prepare technical evidence for review and escalate higher risk use cases through established channels.
Support investigators through consultations, demonstrations, training materials, and reusable examples; contribute technical content to publications, grants, and multi site research collaborations.
Track service measures such as request turnaround, cohort or feasibility output quality, reuse of methods, model performance, and investigator feedback; report progress and operational issues.