SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Viz.ai is seeking a Clinical Data Scientist to lead Evidence Generation data strategy and execution. This role combines clinical research data management expertise with hands-on data engineering and statistical analysis capabilities.
You will own the full data lifecycle for clinical research and quality improvement studies: from raw EHR extraction through database build, cleaning, transformation, QC, and data lock. Key responsibilities include:
**Research Database & Data Management:** Configure and maintain EDC forms and eCRF specifications; build, validate, and maintain research databases end-to-end; develop and run data cleaning workflows with audit trails and inspection-ready documentation.
**EHR Data Transformation & Pipeline Engineering:** Pull and parse raw EHR data (FHIR bundles, HL7v2 messages, flat-file exports) from site and enterprise sources; design and build transformation pipelines in Python/SQL that clean, standardize, and reshape raw EHR extracts into structured, validated, EDC-ready datasets; establish version-controlled, repeatable transformation code (not one-off scripts) that can reliably re-run as new extracts arrive across sites and studies.
**EHR-to-EDC Automation & Data Model Strategy:** Design and operate data model strategy for automated ingestion of EHR-level real-world evidence into the EDC; work hands-on with APIs/integration endpoints; unify disparate data models accounting for site/EHR variability, mapping logic, versioning, and schema evolution.
**EHR Systems Fluency:** Partner with site IT/informatics and Product teams to understand EHR constraints, extract structures, and change management; translate EHR data realities into feasible study data capture and monitoring plans.
**Documentation & Enablement:** Create and maintain DMPs, SAPs, SOPs, runbooks, transformation/mapping specifications, and training materials that operationalize processes across care pathways.
**Data Science & Statistical Analysis:** Perform statistical analyses on cleaned datasets (descriptive, comparative, time-to-event); apply data science techniques (feature/cohort derivation, exploratory data analysis, data quality scoring) to accelerate study readiness and reduce manual review burden; support evidence packages and reporting.
Viz.ai is the leader in AI-powered Care Pathways, deployed in 2,000 US hospitals and trusted by leading life sciences companies. The platform combines real-time multimodal clinical data with clinician engagement to detect disease earlier, coordinate care teams, and ensure faster treatment. Viz.ai was the first company awarded CMS reimbursement for AI and is ranked #1 Healthcare AI Platform by hospitals in Black Book Research.
This is a remote-first role with flexibility across the US and EU. The company offers competitive cash compensation, equity, and comprehensive benefits.
**REQUIREMENTS:**
Required:
- 5–7+ years in clinical research/RWE data management with hands-on database build, cleaning, and QC ownership
- SQL + Python (or R) for data transformation, ETL pipeline development, QC checks, and reproducible pipelines
- Direct, hands-on experience transforming raw EHR extracts (FHIR, HL7v2, or flat-file exports) into structured, analysis/EDC-ready datasets
- Experience with EHR or EHR-derived datasets and understanding of common structures/coding systems (ICD-10, CPT, LOINC, RxNorm preferred)
- Practical familiarity with API-based ingestion and integrating multiple data sources/models
- Experience building/owning DMPs/SOPs/runbooks and maintaining audit-ready documentation
- Working knowledge of version control and reproducible-pipeline practices (e.g., Git) to keep transformation code auditable
- Familiarity with integrating leading AI techniques into work product
Preferred:
- Experience with EDC platforms (Medidata Rave, REDCap, Castor, Veeva, etc.)
- CDISC familiarity (SDTM/ADaM) or strong equivalent standardization experience
- Stats experience in real-world/implementation studies (propensity methods, time-to-event, mixed models)
- Experience with pipeline orchestration/data engineering tooling and cloud (AWS) data warehouses
