Resume

• Authored FIP-37 (Native RoaringBitmap Integration), accepted by community vote, shipped in Apache Fluss 1.0.
• Delivered 12 RoaringBitmap SQL functions natively in FlussCatalog, removing the external UDF JAR dependency.
• Fixed a priority=critical catalog validation bug blocking the Apache Fluss 1.0 release.
• Adopted in production analytics pipelines at Alibaba, Ant Group, Cisco Webex, and Fresha.

• Built 15+ REST API endpoints and JWT-based authentication flows using Java and Spring Boot for a FinTech payment reconciliation platform handling 5k+ daily records.
• Designed relational schemas and integrated backend services end-to-end with Spring Data JPA and PostgreSQL.
• Implemented transactional validation workflows and database query optimizations, ensuring strict data consistency and sub-second transaction lookups.

My flagship contribution: designing bitmap-native analytics for Apache Fluss, a distributed streaming storage system under the Apache Software Foundation.

Distributed rate limiter in Java implementing Token Bucket and Sliding Window algorithms with thread-safe locking, Redis Lua scripting, and Spring Boot REST APIs.

Job application tracker REST API built with Java 21 and Spring Boot 3, featuring state-machine status transitions, JWT authentication with per-user data isolation, Flyway migrations, and MySQL.

Extension library implementing native bitmap analytics functions for Apache Flink and Apache Fluss, designed following Fluss Improvement Proposal [FIP-37] aggregate bitmap functions, set operations, cardinality queries, and a serialization layer.

Production-style full-stack online bookstore with role-based auth, cart, orders, inventory and an admin dashboard, backed by REST APIs on MongoDB.

Cloud-native microservices application with API Gateway, Eureka discovery, JWT auth and pagination — a decomposed backend architecture running on Docker.

AI-assisted fitness platform generating personalized workout and nutrition plans with BMI analysis and diet recommendations.