๐ Born in Wenzhou, China | ๐ป Amateur Developer & Environmental Sciences Student | ๐งช Member of PKU LAB409
- ๐ Student at College of Environmental Sciences and Engineering, Peking University
- ๐ ๏ธ Interests: Python Utilities, AI/LLM Applications, and AI4S
- ๐ฑ Currently exploring LLM-powered game platforms and applying AI into environmental sciences and engineering
- In particular, atmospheric envinronmental physics and chemistry, atmospheric environmental monitoring
My proudest "masterpiece" with LAB409.
A next-gen online Avalon AI battle platform, blending classic social deduction with LLM intelligence.
Highlights:
- Upload your own AI strategy and battle against others
- Integrated LLM decision helper for smarter and more human-like agents
- Game replay, rankings, and full-featured admin tools
- Secure, scalable, and modular architecture
Tech Stack:
Python (Flask, SQLAlchemy) ยท HTML/Bootstrap ยท JavaScript ยท OpenAI API
Automating one of the most painful parts of academic writing.
An AI-powered document formatting agent that converts messy drafts into clean, standardized Word documents.
Highlights:
- ๐ Automatically formats headings, fonts, spacing, and layout
- ๐ Fixes inconsistent styles across the document
- ๐ฏ Adapts to academic / report formatting requirements
- ๐ง Structure-aware (abstract, sections, figures, tables)
- โก Saves massive manual formatting time
Tech Stack:
Python ยท python-docx ยท LLM APIs ยท rule-based + AI hybrid pipeline
Motivation:
Academic formatting is repetitive, error-prone, and time-consuming.
This project focuses on AI for real productivity, not just demos.
Future:
- LaTeX โ Word conversion
- GUI / plugin integration (Office / Overleaf)
- Custom template system
My personal site โ a playground for design, experiments, and showcasing my works.
- Languages: Python, C++, R, HTML, CSS, JS/TS
- Frameworks: Flask, Bootstrap, FastAPI, Vue.js, uni-app
- Databases: SQLite, SQLAlchemy
- ML & AI & LLM: Regression / classification models, OpenAI API, custom AI agents
- Other: Git, LaTeX
- Expanding Tuvalon with new features & better AI safety
- Discovering AI for (environmental) sciences (like PIML, Symbolic Regression)
I'm open to:
- ๐ก Interesting AI/LLM experiments
- ๐ง Tools that make life easier
Reach me via email: dmcnczy@163.com
"I taught AI to lie in Avalon. Now I canโt trust it anymore."