Skip to content
View WW-shan's full-sized avatar

Highlights

  • Pro

Block or report WW-shan

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
WW-shan/README.md

WW

Quant Research · AI Systems · Backend / Full-Stack
量化研究 · AI 系统 · 后端 / 全栈开发

📧 212500581@qq.com  ·  GitHub

Python FastAPI TypeScript React Go Docker


👋 About Me / 关于我

我喜欢把一个模糊的想法拆成可以运行、可以检查、也可以继续迭代的东西:先弄清楚问题和数据,再把实验做成代码,最后整理成别人能用的工具。

I like turning vague ideas into things that can be run, checked, and improved: understand the question and the data first, turn the experiment into code, then package it into something useful.

目前主要在做三类项目:

  • Research systems / 研究系统 — market data, validation, replay, backtests, reports
  • AI & backend tools / AI 与后端工具 — agents, APIs, workers, dashboards, automation
  • Small products / 小型产品 — study tools, bots, and practical web applications

🎯 正在寻找量化、AI 或后端开发实习。

Currently looking for a Quant, AI, or Backend internship.


🧭 How I Work / 我的做事方式

Question → Data → Checks → Experiment → Review → Ship
问题    → 数据 → 校验  → 实验    → 复盘  → 上线

我更在意过程是否清楚、结果能否复现,以及失败时能不能知道问题出在哪里,而不只是展示一张漂亮的结果图。

I care about a clear process, reproducible results, and knowing why something failed—not just showing a pretty chart.


🎨 Stack / 技术栈

Area Tools 方向
Core Python · TypeScript / JavaScript · Go · Java 后端、脚本、全栈项目
Research Market data · Backtesting · CatBoost · BC / PPO 数据研究、策略实验、模型验证
Backend FastAPI · Workers · REST APIs · Telegram bots API、后台任务、消息推送
Frontend & Delivery React · Vite · Docker Compose · GitHub Actions · GitHub Pages 控制台、部署、CI

🎞️ A Research Loop in 9 Frames / 九格研究小故事

这不是我的真实日常照片,而是把我做项目时常见的一条路径画成了一个小故事:发现问题 → 清理数据 → 质疑结果 → 做成工具 → 留下下一个问题。

This is not a literal diary. It is a visual metaphor for a typical project loop: find a problem → clean the data → challenge the result → build a tool → find the next question.

Discover a mismatch
01 · Discover / 发现
两份图表对不上。
Circle the mismatch
02 · Trace / 追查
先找到错位的位置。
Align timestamps
03 · Align / 对齐
把时间和数据重新理顺。
Clean the data
04 · Clean / 清理
让输入先变得可信。
Question a backtest
05 · Stress-test / 质疑
结果太漂亮,反而要小心。
Add costs to the experiment
06 · Rebuild / 重做
把成本和现实条件放进来。
Compare experiments
07 · Compare / 比较
看清哪个结果值得继续。
Turn research into a tool
08 · Build / 做成工具
把研究变成能使用的界面。
Find the next question
09 · Iterate / 继续
一个答案,通常会带来下一个问题。

⭐ Selected Work / 精选项目

Project What it shows / 项目亮点 Stack
poly_strategy Prediction-market research pipeline: collection, relation discovery, replay, backtest and dry-run checks. / 预测市场研究链路:采集、关系发现、回放、回测和干跑验证。 Python · Polymarket · Kalshi
Atlas20 Point-in-time rotation research with reproducible backtests, a FastAPI worker and a React/Vite console. / 可复现的轮动回测、FastAPI worker 和 React/Vite 控制台。 Python · FastAPI · React · Docker
Crypto_Research_Agent An autonomous research workflow with public data, validation, reports and paper-simulated experiments. / 用公开数据做验证、报告和纸面实验的研究 Agent。 Python · Agents · SQLite
crypto-alpha-portfolio A structured way to test token-unlock, wallet, funding and anti-Sybil hypotheses before putting them into a portfolio. / 系统研究 token unlock、钱包行为、资金费率和反女巫假设。 Python · Research · Backtesting
meme BSC token lifecycle data → dataset → hybrid model → bot, with explicit paper/live safety boundaries. / 从链上生命周期数据到数据集、混合模型和 bot,并明确区分 paper / live。 Python · BSC · CatBoost · RL
LTT_Strategy A multi-timeframe signal monitor using Bitget/Yahoo data and Telegram alerts. / 多周期信号监控,连接 Bitget、Yahoo Finance 和 Telegram。 Python · Telegram · APIs

🗂️ More Projects / 更多项目

  • ai-practice-study — Interactive AI revision site with 51 topics and 130 questions. / 51 个考点、130 道题的 AI 复习站。
  • polyFIFA2026 — Football prediction research prototype. / 足球预测研究原型。
  • ArenaFIFA2026 — Football data research project. / 足球赛事数据研究项目。
  • fundingRate — Funding-rate arbitrage research. / 资金费率策略研究。
  • second-hand-trading — Java full-stack second-hand marketplace project. / Java 全栈二手交易平台。
  • KXO325 — Bilingual logistics study guide with a live demo. / 带在线演示的双语物流复习站。

📬 Contact / 联系我

如果你在招 量化、AI 或后端开发实习,欢迎直接发邮件:

If you are hiring for a Quant, AI, or Backend internship, email me directly:

📧 212500581@qq.com

也欢迎通过 GitHub 联系我:

You can also reach me through GitHub.


Thanks for stopping by. / 谢谢你看到这里。

Pinned Loading

  1. four_meme four_meme Public

    BSC token lifecycle collector, dataset builder, hybrid model trainer, and research bot runtime.

    Python 2

  2. Atlas20 Atlas20 Public

    Atlas20 Rotation research framework and web console for crypto rotation backtesting

    Python 1

  3. crypto-alpha-portfolio crypto-alpha-portfolio Public

    Structured crypto alpha research across token unlocks, wallet behavior, funding, and anti-Sybil signals.

    Python 1

  4. Crypto_Research_Agent Crypto_Research_Agent Public

    Autonomous crypto research agent for on-chain signals, data validation, reports, and paper-traded experiments.

    Python 1

  5. poly_strategy poly_strategy Public

    Prediction-market research toolkit for collection, replay, and dry-run validation.

    Python 3 1

  6. LTT_Strategy LTT_Strategy Public

    Multi-timeframe crypto signal monitor with Bitget/Yahoo data and Telegram alerts.

    Python 2