I build systems that turn messy healthcare workflows into reliable, automated, data-driven operations.

Operational reporting depended on fragile interpretations of clinical events, configuration rules, and downstream stakeholder expectations.

Healthcare data had to remain explainable, auditable, and aligned with how clinicians actually document inside Epic.

Modeled the data flow as a pipeline: source events, business rules, validation checks, exception handling, and stakeholder-ready output definitions.

Created a stronger blueprint for analytics work that reduces manual reconciliation and makes reporting assumptions explicit.

Multiple anesthesia profiles carried inconsistent build patterns, making support harder and increasing the risk of charging variation.

The work had to respect operational differences while avoiding one-off logic that would become impossible to maintain.

Compared profiles, identified repeatable build patterns, documented exceptions, and pushed toward a cleaner standard configuration model.

Improved maintainability and gave stakeholders a clearer path for future billing logic changes.

Manual review of perioperative logic created friction, inconsistent interpretation, and unnecessary dependency on tribal knowledge.

Clinical, billing, and compliance-sensitive logic required careful validation and clear ownership.

Mapped decision paths, isolated rule conditions, identified repeatable checks, and shaped automation concepts around human-reviewable outputs.

Reduced ambiguity around operational logic and created a foundation for safer automation prototypes.

Reporting workflows needed to support many stakeholders without depending on repetitive manual execution.

Batch output had to remain stable, understandable, and easy for operational consumers to trust.

Reframed recurring reports as scheduled, native batch processes with clearer ownership, timing, and validation expectations.

Improved repeatability and made the reporting workflow easier to scale across consumers.

OR readiness metrics only help when the data can answer who, what, when, and why without starting a blame game.

Analytics needed to be fair, clinically credible, and sensitive to real perioperative workflow complexity.

Explored attribution logic, event timing, readiness definitions, and operational views that make bottlenecks visible.

A stronger analytic frame for improving perioperative throughput and conversation quality.

Privilege reviews are often detail-heavy, repetitive, and easy to slow down when data lives across systems.

Any automation must preserve reviewability, traceability, and human approval for sensitive provider data.

Outlined an audit workflow that compares source records, flags mismatches, and produces reviewer-friendly exception lists.

Created a pragmatic automation concept for reducing review burden while keeping governance intact.

Provider bio and SER-related cleanup work can become repetitive, manual, and vulnerable to inconsistent source formatting.

Imported data needs normalization, validation, and clear separation between automated suggestions and approved updates.

Prototyped a scraper/import pattern with source extraction, field mapping, cleanup rules, and review checkpoints.

Established a reusable pattern for data cleanup projects where automation accelerates the work but does not bypass stewardship.