Data science · ML · Engineering
I work on clinical-trial analytics at Amgen and consult through Epsilon DS on ML, data, and software.
Pour-over first, then pull requests.
For ten years I have built systems that have to survive contact with the real world: predictive models for a global insurer, clinical-trial analytics at one of the largest biotech companies, and software that small businesses run on every day.
I am a mathematician by training and an engineer by practice. I care about dataset lineage, interfaces that concede to their users, and the version-controlled history of a decision.
By day I am on staff at Amgen, on the Operational Design Analytics team, planning the next generation of clinical trials. Alongside, and kept separate, I run Epsilon DS: a small studio for senior, hands-on machine learning, data engineering, and bespoke software.
I am part of the team that turns Amgen's historical and operational data into the foundation for planning its next generation of clinical trials.
Surface feasibility parameters and historical trial performance metrics (enrollment, screen-failure, and drop-out rates) to inform study design.
Operate predictive analytics tooling (Trial Trove, Site Trove, DQS) to advise placement and planning decisions across therapeutic areas.
Collate global site performance data and recommend geographic footprints calibrated to each study's specific requirements.
Curate historical reference data into high-confidence modeling sets; serve as a primary steward of dataset lineage and quality.
NDA work is described by shape only. Full case studies on request.
| Area | Scope | Stack |
|---|---|---|
| Data science and ML | Production machine learning, statistical modeling, NLP, computer vision, and forecasting. Built and shipped end-to-end at enterprise scale. | Python, R, PyTorch, TensorFlow, Polars, Snowpark |
| Data engineering and MLOps | Pipelines, warehouses, observability, and the unfashionable plumbing that turns models into systems people actually depend on. | Snowflake, Docker, Kubernetes, Airflow, MLflow, DBT |
| Web and application development | Next.js applications, dashboards, internal tools, point-of-sale systems, and cross-platform mobile work. | Next.js, TypeScript, Flutter, Swift, Flask, Django |
| Advisory | Engagements as a fractional lead, technical reviewer, or hands-on contributor where the project deserves senior judgment. | Architecture, hiring, roadmaps, code review, data audits |
The consulting studio I operate alongside the day work. Engagements span machine learning, data engineering, and bespoke software, from clinical pipelines to point-of-sale platforms.
The studio exists to give ambitious teams senior, hands-on data and software work that is otherwise difficult to find. Every engagement is conducted by the principal. When extra hands are needed, those hands are vetted personally.
2021-2023
Master of Information & Data Science, Berkeley, CA
Rigorous, interdisciplinary program spanning machine learning, statistical inference, and information systems, taught largely through real datasets and industry partnerships.
Capstone
A World for Every Child →
Machine learning to support early childhood-cancer diagnosis in under-resourced regions.
2011-2015
Bachelor of Science, Mathematics, Montclair, NJ
Foundations in analysis, algebra, statistics, and computational methods. The mathematical undercarriage on which the rest of the work rests.
