Payments & Checkout Systems

• Designed and built highly scalable, low-latency distributed services integrated into Amazon’s checkout flow, handling tens of thousands of transactions per second (TPS) with response times in tens of milliseconds.

• Orchestrated interactions across dozens of dependency services supporting checkout experience, payment method selection, payment plans, order processing, fulfillment, and transaction reconciliation.

• Architected resilient service-to-service communication patterns to ensure high availability and fault tolerance in mission-critical payment workflows.

Gamification Offer Platform

• Designed and developed an event-driven Gamification Offer Platform enabling business teams to launch dynamic promotional campaigns tied to customer actions.

• Built distributed components for customer targeting, state management, event ingestion, reward issuance, batch processing, and notification workflows.

• Implemented scalable event processing pipelines to support high-throughput behavioral tracking and real-time offer activation.

Prime Benefits Aggregation Layer

• Developed an orchestration and aggregation layer to consolidate Amazon Prime customer benefits from dozens of upstream data sources.

• Built asynchronous read and cache update mechanisms to support thousands of TPS with sub-50ms latency requirements.

• Designed pre-computation services and materialized view caching strategies to enable single-digit millisecond data retrieval.

Third-Party Payment Integration

• Led end-to-end integration with external payment partners to enable third-party payment experiences on Amazon.

• Designed secure APIs and authentication flows (OAuth, multi-factor authentication) in collaboration with partner engineering teams.

• Implemented end-to-end encryption and tokenization mechanisms to securely handle payment and PII data.

• Delivered production systems serving tens of millions of daily active users (DAU).

• Design and develop new features for web application using JavaScript and React framework.

• Build and maintain stable API gateway, provide efficient data query methods, largely reduced number of requests and application/web pages loading time.

• Process and manage data for NLP predict model and interactive data visualization, introduce visualization module like 'Theme exploration', 'Focus Insight' to integrate model predict result with our platform visualization tools.

• Integrate third-party software(e.g. intercom, slack...)and build data connectors with other data analytic platform(e.g. Microsoft PowerBI...)

Teaching Assistant of CSEE 4119 Computer Networks

• Built Data connectors between our platform and Microsoft PowerBI. Including designing the user interface, data packaging and processing, data communication through PowerBI restful API and enabling Oauth Authentication through Microsoft ADAL and MSLA.

• Upgraded frontend D3 library from v3 to v5 and introduced Webpack to conduct file management, decreasing page loading time by 20%

• Introduced a map visualization module to our platform using MapBox GL, enabled our platform to display geometric data and view them in different view modes(heat-map/user-define marker layer/svg regional map)

• Designed and Implemented new visualizations to visualize data in cluster and hierarchy structure. Enriched our frontend visualization library to meet our customer’s requirements, using JavaScript and React.

• Came up with a new control strategy named 'Three-Segment' to control the DC-DC converter in electric vehicle charger to improve efficiency

• Simulated different 'Three-Segment' control strategies in MATLAB

• Compared the loss and efficiency of different 'Three-Segment' control strategies and concluded one with highest efficiency in MATHEMATICA