GitHub-Pages

CHEN Yanshao | 陈沿劭

Aspiring Data Scientist & Data Analyst | Generative AI & Agentic RAG Developer

Top Skills:

🙋 About Me

I am a Master of Science in Data Science student at Lingnan University with a background in Information Management. I specialize in developing Agentic RAG frameworks and modular AI pipelines to solve complex data challenges. With over two years of professional experience as a Management Trainee, I excel at bridging the gap between technical AI capabilities and scalable business solutions.

Education: MSc in Data Science, Lingnan University (Expected Aug 2026)
Strengths: Cross-departmental project leadership, technical documentation, and AI solution development.
Focus Areas: LLM Orchestration, Prompt Engineering, and Automated Data Synthesis.

🛠 Skills

Technical: Python (Pandas, Numpy, Scikit-learn), SQL, R, Machine Learning, Prompt Engineering.
Business: Cross-functional Project Management, Process Optimization, Technical Documentation, Metric Tracking.
Tools: Google Gemini API, Vertex AI, Power BI, MySQL, MongoDB, Git, AutoCAD.

🚀 Projects (Portfolio)

[Project 1] T2P: Term 2 Project - Modular Mathematics Analytics System (Jan 2026 - July 2026)

Problem: Scarcity of high-quality, logically rigorous calculus datasets and structured analytics for tracking higher mathematics mastery.
Data: Synthetic student performance logs and structured JSON datasets containing step-by-step mathematical solutions generated via Gemini 2.5 Pro.
Approach:
- Built a multi-agent orchestration flow in Python to automate educational data synthesis.
- Developed an Agentic RAG architecture and a Quality Evaluation Module to ensure pedagogical consistency.
Outcome: Created a reproducible AI pipeline and a functional Learning Analytics Dashboard that tracks 15+ topics against an 85% Mastery Target.
Contribution: Sole Developer. Designed the entire system architecture, from prompt engineering to the final data visualization dashboard.

Fig 1. T2P System Output: Radar chart of Knowledge Mastery and Topic Progression Status.

[Project 2] E-commerce SQL vs. NoSQL Performance Study (Oct 2025 - Nov 2025)

Problem: Identifying optimal database architectures for high-concurrency e-commerce scenarios to improve system latency.
Data: Simulated transaction data covering user accounts, products, and inventory, modeled via comprehensive ER diagrams.
Approach:
- Designed a High-Availability System Architecture featuring a Load Balancer, Cache Layer (Redis), and Master-Slave replication.
- Developed a detailed Relational Data Model (ERD) to manage complex e-commerce entities and transaction flows.
- Executed benchmarking on MySQL and MongoDB to compare CRUD performance and complex query handling.
Outcome: Produced a technical report providing data-driven recommendations for database selection based on quantitative performance metrics.
Contribution: Led the experimental design, executed performance stress tests, and created all technical system diagrams and ER models.

Fig 2. E-commerce System Architecture and Entity-Relationship Diagram (ERD) designed for performance benchmarking.

📄 Contact & Resume

Email: 17796163646@163.com
Resume: Download My Full CV (PDF)
Links: GitHub Profile LinkedIn