AI Systems · Model Research · Full-stack Engineering
Slow is fast. Build openly. Let small changes compound.
Undergraduate student in Data Science and Big Data Technology at Shanghai University of Electric Power.
- Agent systems: LangChain, LangGraph, MCP, RAG, tool security, and production backends.
- Model research: dynamic MoE routing, LoRA and model safety, long-term memory, and visual tokenization.
- Quantitative finance (in progress): exploring market data, technical indicators, agent-assisted analysis, and QMT workflows.
- Product engineering: clinical AI, React Native applications, and Python/Windows developer tools.
| Area | Projects |
|---|---|
| Financial Data & Quant Research | FinanceMCP · TradingAgents-MCPmode · QMT-MCP · bigqmt-history-export |
| Agent Safety | tool-output-spoofing-lab |
| Model Research | qwen3.8-research · Dynamic MoE Notes · Qwen LoRA · LLM Memory · Image Tokenization |
| Product | littletask · Dolphin ClinReviewPlatform |
| Developer Tools | md2docx · Tencent Meeting AI Notes |
Languages
AI & Agents
Application Engineering
Data & Quant Tools
Storage & Infrastructure
Slow is fast. I want to use open source to improve myself first, then help change the people and systems around me. Step by step, I hope those small changes can grow into a wider ecosystem, broaden shared understanding, and make a meaningful contribution to human progress.