Algorithm Engineer @ JD.com | 4 yrs exp as Advertising Algorithm Engineer (2021-2025) | Now i am LLM Agent Designer
Stack: PyTorch, TensorFlow, Python, JavaScript
- 💻 Checkout my website here.
I build practical AI systems, LLM agent workflows, quantitative research tools, and algorithm prototypes. My repositories are mostly small, self-contained projects that turn an idea into a runnable tool.
| Project | Focus | Notes |
|---|---|---|
| love-stock | Stock analysis agent toolkit | Standalone A-share, US stock, and HK stock analysis workflow. |
| autoquant | Quant research | Python toolkit for market data, factor research, and trading analysis. |
| AgentRelay | Agent infrastructure | Lightweight relay utility for agent and developer workflows. |
| agent-watcher | Mobile companion | Apple Watch, iPhone, and iPad companion for Claude Code and Codex. |
| Image-Retrieval-With-User-Feedback | ML retrieval | VGG16 + SVM image retrieval with user feedback. |
| mimir | AI experiment | Python project for LLM/data workflow experiments. |
| Repository | Description |
|---|---|
| goal-refiner-skill | Skill for refining vague goals into executable task briefs. |
| earnings-watch-skill | Earnings and filing monitor skill. |
| notion-publisher-skill | Notion publishing workflow skill. |
| follow-github | Personalized GitHub digest for people, trends, and hot repositories. |
| oh-my-stock | Lightweight stock analysis workflow. |
| AgentRelay | Python relay utility for agent workflows. |
| Repository | Description |
|---|---|
| autoquant | Quantitative research and stock-analysis toolkit. |
| love-stock | Deep stock-analysis agent toolkit. |
| lof-arbitrage | China LOF discount/premium arbitrage monitor with reports and email alerts. |
| Tencent_2019_ad_algorithm_competition | Baseline for Tencent 2019 advertising algorithm competition. |
| Repository | Description |
|---|---|
| ACE_algorithm | Alternating Conditional Expectations implementation. |
| BP_sin | Backpropagation demo for fitting a sine function. |
| DecisionTree | ID3 decision tree implementation. |
| NB_for_Spam_Filtering | Naive Bayes spam filtering. |
| Quadratic-Discriminant-Analysis | Python implementation of Quadratic Discriminant Analysis. |
| RL4AUCTION | Reinforcement learning / auction mechanism experiment. |
| SoftMaxRegression | PyTorch softmax regression exercise. |
| Image-Retrieval-With-User-Feedback | Image retrieval with VGG16, SVM, and user feedback. |
| Repository | Description |
|---|---|
| agent-watcher | Apple Watch/iOS companion for coding agents. |
| MikiZoo | Swift app experiment. |
| evercomment | Rust comment/workflow experiment. |
| ds_hotmap | TypeScript hotmap interface. |
| weixin-miniprogram | WeChat mini-program project. |
| resume | TeX resume template. |
| Repository | Description |
|---|---|
| mimir | Python AI/data experiment. |
| blog_comment | Blog comment storage repository. |
| clash_file | Small networking configuration/archive repository. |
| get-pip | Small setup/archive repository. |
| Miraclemin | GitHub profile README repository. |
- LLM agents and coding-agent workflows
- Quantitative research and financial-data automation
- Advertising algorithms and recommender systems
- Practical ML systems that can be shipped, monitored, and reused