Research
Decision Architect · Harvard PhD · Former faculty (Operations/Tech) · Founder & CEO of Doogooda (Decision Systems)
I study how institutions make decisions under uncertainty—where constraints, governance, and trade-offs matter as much as prediction accuracy.
I design decision systems that remain defensible under audit—where assumptions, constraints, and trade-offs are explicit.
My work bridges decision science and deployment, focusing on auditable decision systems for regulated operations in healthcare, public policy, and AI governance.
Method: Causal inference → scenario simulation → constrained optimization → auditable decision artifacts.
Selected Publications
Healthcare Operations · Production and Operations Management · 2024
Facility Closure Decisions Under Regulatory Constraints: A Causal Inference Approach
How to structure facility closure decisions when regulatory constraints bind. Framework adopted by US academic medical centers.
Why it matters: Changes how hospitals make facility closure decisions—from political negotiation to evidence-based, defensible choices.
AI Governance · Conference Proceedings · 2024
Autonomous Systems Require Governance-by-Design: An Accountability Framework
Moving beyond explainability to institutional accountability. Governance structures for autonomous decision systems.
Why it matters: Defines who approves AI decisions in defense and public sectors—closing accountability gaps before deployment.
Decision Science · Working Paper · 2024
Decision Packets Over Recommendations: Artifacts for Institutional Approval
Why institutions need decision artifacts—not just "recommendations"—with explicit assumptions, constraints, and trade-offs.
Why it matters: Shifts AI from "black box outputs" to reviewable decision memos that executives can actually approve or reject.
Public Policy · Government Report · 2024
AI-Driven School Facility Optimization: Evidence for Gyeonggi Provincial Council
Causal demand forecasting and constraint-based optimization for public sector resource allocation.
Why it matters: Enables legislators to justify school consolidation with data—reducing political gridlock in education planning.
Healthcare Policy · Dissertation · 2020
Causal Inference for Healthcare Operations Under Policy Constraints
Methodological framework for decision-making in regulated healthcare contexts. Harvard PhD dissertation.
Working Papers & In Progress
Theme 1: Decision Intelligence for Regulated Operations
Question: How to operationalize decision systems when constraints and governance requirements matter as much as accuracy?
Output: Co-authored paper with industry practitioners · Framework validation in healthcare pilot
Theme 2: Governance-by-Design for Autonomous Systems
Question: Where do accountability failures occur in deployed AI systems, and how can governance structures prevent them?
Output: Case study series (defense, healthcare, public sector) · Governance design patterns brief
Theme 3: Evidence Standards in Policy Narratives
Question: How to evaluate causal claims in political and policy discourse where controlled experiments are impossible?
Output: Methodological framework paper · Public-facing evidence checklist
Data/Methods Notes
Auditability First
All analysis includes explicit assumptions, data lineage, and sensitivity checks. No hidden degrees of freedom.
Assumption Transparency
I label what's empirically verified vs. what's assumed. Trade-off tables show what changes if key assumptions shift.
Reproducibility
Methods documented with sufficient detail for replication. Code and data availability specified in each publication.
Constraint-First Modeling
Real institutions operate under binding constraints (regulatory, operational, political). Models that ignore constraints aren't deployable.
Academic Background: Harvard PhD (Decision Science, Health Policy) · Yale MA (Statistics) · Caltech BS (Applied Math)
Affiliations: Visiting Researcher, Seoul National University Cognitive AI Lab · Former Assistant Professor, UCL School of Management
Full Credentials
Collaboration inquiries → Email