Skip to content
View b1ue13e's full-sized avatar
🏠
Working from home
🏠
Working from home
  • Anhui University of Finance and Economics
  • Bengbu, China

Block or report b1ue13e

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.

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
B1ue13E/README.md

Junyu Li - AI Systems and Quant Research

Hi, I am Junyu Li

Statistics Undergrad @ Anhui University of Finance and Economics
AI systems, time-series forecasting, sparse MoE, quantitative research, and product-grade research tooling.

GitHub Email Focus Location


Profile

I build research and product systems for noisy, non-stationary real-world signals. My current work sits at the intersection of statistical learning, sparse neural architectures, market data, and decision-oriented software.

I care about models that survive contact with messy data: robust forecasting under corrupted SCADA, alpha-oriented signal extraction, practical evaluation protocols, and tools that turn scattered information into repeatable workflows.

Current Direction

Track What I am working on
AI for time series Sparse MoE, robust forecasting, edge-aware model design, corrupted sensor data
Quant research Market regime modeling, signal extraction, high-noise financial time series
Research tooling Local-first learning tools, API hubs, workflow systems, evaluation dashboards
Product engineering Mobile tools, decision-support interfaces, polished prototypes

Selected Work

Project Focus Stack
Ultra-LSNT Sparse MoE framework for wind power forecasting under corrupted SCADA and edge deployment constraints Python
kronos-hub Integration-first API hub for forecasting, trading research, and AI hedge-fund style workflows Python
LoveACE Campus-life mobile toolbox for Anhui University of Finance and Economics Dart
fengshui-lens Product-style decision app for Shanghai rent, city-settlement, and real-world life training TypeScript
TrendRadar AI-assisted trend monitoring and information analysis toolkit Python

Toolbox

Python TypeScript Dart JavaScript PyTorch Pandas Git Linux

Research Taste

  • Prefer mechanism-aware models over leaderboard-only improvements.
  • Prefer small, reproducible validation gates over vague claims.
  • Prefer local-first tools when privacy, iteration speed, or offline use matters.
  • Prefer interfaces that make hard thinking easier instead of merely looking busy.

GitHub Analytics

GitHub stats Top languages

GitHub streak

Contact

I am open to conversations around AI for time series, quantitative research, sparse modeling, and research-to-product systems.

Popular repositories Loading

  1. Ultra-LSNT Ultra-LSNT Public

    Official implementation of "Wind power forecasting under corrupted SCADA for edge deployment: sparse MoE with feasibility guidance". Featuring Ultra-LSNT and FG-MoE.

    Python 1

  2. fengshui-lens fengshui-lens Public

    A product-style open source app for Shanghai rent decisions, city-settlement handoff, and real-world life training.

    TypeScript 1

  3. athlete-insight athlete-insight Public

    TypeScript 1

  4. - - Public

    Python 1

  5. kronos-hub kronos-hub Public

    Integration-first API hub for Kronos forecasting, TradingAgents research, and AI Hedge Fund execution/backtesting.

    Python 1

  6. TrendRadar TrendRadar Public

    Forked from sansan0/TrendRadar

    🎯 告别信息过载,AI 助你看懂新闻资讯热点,简单的舆情监控分析 - 多平台热点聚合+基于 MCP 的AI分析工具。监控35个平台(抖音、知乎、B站、华尔街见闻、财联社等),智能筛选+自动推送+AI对话分析(用自然语言深度挖掘新闻:趋势追踪、情感分析、相似检索等13种工具)。支持企业微信/飞书/钉钉/Telegram/邮件/ntfy推送,30秒网页部署,1分钟手机通知,无需编程。支持Dock…

    Python