Shuo Yan Email & Phone Number

360 Wenku (Document Library) Data Migration and Filtration with LLMBeijing, China· Developed a unified document migration system for the 360 document library in collaboration with five document library companies. Leveraged Go language multi-threading, the system was deployed on Kubernetes, featuring no-code automatic migration and statistical tools. This approach enabled a successful migration of around 5 million new pieces of document data.· Coordinated with Microsoft, implementing 360 Smart Brain LLM to streamline content filtration of the document library. Provided Microsoft with high-quality and accurately-filtered Chinese data. Automated supply of 50,000 high-grade Chinese documents to Microsoft weekly.· Constructed an API interface to transform document content from the 360 document library into corpuses usable for model development. Enhanced data usability and quality for the training of the 360 large model.· Utilizing the company's proprietary XJSON data structure storage, integrated with SPDK, the initiative significantly raised corpus access speed - marking a 3.2 times improvement compared to using only the JSON format.

Non-Profit Recycling Management SystemInglewood, CA· Developed and optimized an order management system based on Magento 2. Enhanced MySQL database integration led to a 30% increase in data processing efficiency for user activities and bonus allocations.· Configured Apache/Nginx web servers to ensure high availability and efficient handling of peak user traffic periods, and maintained a 99.9% uptime post-launch.· Employed XAMPP for a flexible cross-platform development setup, and delivered substantial enhancements to system performance and compatibility across various user devices.

floodwatch.io, Flood monitoring applicationCharlottesville, VA· Co-authored the paper titled "Validating Crowdsourced Flood Images using Machine Learning and Real-time Weather Data," published in IEEE/IFIP Intl. Conference on BigDataSE-2022.· Accelerated geographic location acquisition by multi-threading and asynchronous IO to achieve large-scale geometric data verification, and successfully improved data acquisition speed by 30 times.· Cooperated with the Vietnam government to set up LoRaWAN flood monitoring network and built a cloud platform with DynamoDB for data collection on AWS IoT, which now covers major cities in Vietnam.

CS4102 - Algorithm· Organized weekly algorithm discussions and algorithm problem analysis and helped develop automatic scoring test cases and grading rubrics for coding homework.

Master, Computer Science And Engineering

Research Interest: Machine Learning, AI, Natural Language Processing, Software Engineering, LLM

Bachelor, Computer Science, 3.9/4.0

Related Coursework: Algorithm, Computer Architecture, Computer Networks, Cybersecurity, Database Systems, Machine Learning, Advanced.