Haysen Lin

Built a behavioral segmentation model driving a projected ~$50M TVL uplift by translating user behavior into insights on retention, risk concentration, and capital allocation. Designed end-to-end analytics frameworks translating market and on-chain signals into actionable dashboards for user behavior, liquidity exposure, and concentration monitoring. Built a data-driven early-warning and monitoring system for Aave & Ethereum, capturing ~85% of historical stress events and supporting risk-aware liquidity and operational decisions.

Architected a momentum-based early-warning system for Aave & Ethereum capturing ~85% of historical redemption stress events. Built real-time dashboards to support LTV and liquidation parameter analysis during volatility. Designed the L1–L4 Behavioral Risk Architecture, translating risk triggers into auditable on-chain monitoring signals to detect liquidation cascades, concentration buildup, and systemic fragility. Developed a behavioral segmentation model (560k+ addresses) to quantify liquidation sensitivity and concentration, informing retention strategies with projected ~$50M TVL uplift and mitigated cluster-level bad debt exposure.

Built scalable validation and exception-handling frameworks supporting $200M+/month in fund flows, improving anomaly triage and execution integrity across high-volume operations. Improved data quality from 81% to 97% and strengthened audit readiness by redesgning validation schemas, reducing operational risk and generating 13% cost savings. Accelerated risk responsiveness by 34% via centralized exception governance and dynamic resource allocation, reducing settlement disruptions during peak periods. Led risk-informed workflow alignment across teams, implementing internal controls that reduced process discrepancies and strengthened operational resilience.