-Developed a Java Spring Boot scheduling system to extract Cryptocurrency data from multiple exchanges, convert it into a custom format, and load it into a MySQL database on AWS RDS.-Enhanced software reliability by identifying potential system vulnerabilities, implementing JUnit Integration and Performance testing, and promoting best practices within the team.-Created system design documentation including data schema, API interface logic, and physical design.-Implemented a CI/CD pipeline using GitHub, Jenkins, and AWS EC2 for efficient code deployment.

Key achievements:-Enhanced the Jubao service by leveraging C++, an object-oriented, feature-based service framework, resulting in a more robust support system for a multitude of vertical strategy needs, including but not limited to rough rank, precise rank, various quality models, and intricate similarity calculations. This development created a more streamlined, efficient, and adaptable service to cater to diverse project requirements.-Participated in developing both online and offline feature engineering pipelines. Through this, I was able to significantly improve the precision and performance of the Search/Recommend feature, leading to more accurate search results and recommendations and hence, a better user experience.-Implemented various technologies to improve the system's efficiency, including Redis for caching, enhancing speed and responsiveness, Kafka for distributed streaming, ensuring high throughput and fault tolerance, and RPC/gRPC services for effective communication. This augmented the system's data-handling capacity and contributed to overall performance enhancements.-Contributed to developing and optimizing high-throughput systems capable of handling up to 800,000 queries per second (QPS). This level of performance ensured seamless service to our vast daily user base, which numbers over 370 million. The experience has given me a deep understanding of building scalable systems capable of withstanding heavy loads while maintaining efficiency.

As a front-end engineer, I was responsible for writing reliable codes and designing relevant frameworks. Key achievements:-Designed and implemented the homepage interface of the Yinchuan economic development zone service platform with vue3 and JavaScript. -Developed and tested a framework for finding companies in areas users circled on the Baidu map with vue3, JavaScript, and the Baidu map platform.

Key Achievements :-helped the open-source project FinRL achieve 8,100stars (top 1%) on Github.-published a poster "Spectral tensor layer for model-parallel deep neural networks.", including Z. Wang, Z. Xia, and X.-Y. Liu, published on the second Workshop on Quantum Tensor Networks in Machine Learning, NeurIPS 2021.-published paper: B. Gort, X. -Y. Liu, Xinghang Sun, Deep Reinforcement Learning for Cryptocurrency Trading: Practical Approach to Address Backtest Overfitting, ACM ICAIF Workshop on Benchmarks for AI in Finance, 2022.