United States
Analytically-driven product analyst and applied mathematician with experience spanning health-tech, clinical research, and computational science. Skilled at translating ambiguous, open-ended problems into rigorous data models, clean analytics infrastructure, and actionable tools. Experienced building end-to-end data pipelines, cost models, and AI evaluation frameworks. Eager to contribute to impactful work at the intersection of data, product, and science.
- Overhauled the per-patient cost model, replacing an outdated legacy model with a data-driven framework that revealed true lifetime costs more than 10x the previous estimate, directly informing all future study pricing strategy. - Built and owned a centralized pipeline visibility tool used by data operations and PMs, enabling real-time SLA monitoring and saving over 10 hours per week previously lost to manual ad hoc investigation and escalations. - Developed a Case Report Form abstraction quality scoring system using dbt models and dashboards to evaluate human and AI performance, directly informing workflow design for a function representing over 50% of study cost, with cost reductions already underway and targeting a 90% cost reduction within 6 months. - Designed and maintained analytics tables in dbt/SQL on BigQuery, establishing a single source of truth for clean, reliable downstream reporting across the organization.
- Published in JINST as part of the CMS ECAL collaboration on test beam trials: "Performance of the front-end electronics of the CMS electromagnetic calorimeter barrel for the High-Luminosity LHC." JINST 21, P01001 (2026). arXiv:2511.17261 - Created C++ test software to evaluate next-generation ECAL readout electronics for the High-Luminosity LHC upgrade, identifying and fixing a soldering issue in pre-production. - Built an automated QR-code-driven board testing workflow using an Oracle/Python pipeline, enabling non-technical operators to log results to a centralized repository and scientific collaborators to analyze performance data at scale, supporting full production throughput validation. - Served as sole on-shift operator in the CMS detector control room during year-end maintenance and upgrades, independently monitoring live detector health and safety systems.
- Independently built image registration program in MATLAB to perform two-photon microscopy image data analysis including de-warping, merging and vectorizing. - Extended and improved 3D image vectorizing software mapping oxygen flow in brains of awake mice.