Back to feed
Dev.to
Dev.to
7/30/2026
The original title is: "Open-Source AI System for Chinese A-Share Stock Analysis: Three-Pillar Approach and Results"

The original title is: "Open-Source AI System for Chinese A-Share Stock Analysis: Three-Pillar Approach and Results"

Original: I Built an AI That Analyzes Chinese A-Share Stocks Using a Three-Pillar System — Here's What It Caught Today

Short summary

An open-sourced AI system analyzes Chinese A-share stocks across three pillars: technical indicators (MACD, KDJ, RSI, Bollinger Bands), fundamental screening (DuPont ROE, PEG, Buffett ratios), and sentiment/capital-flow tracking. Stocks scoring 12+ out of 30 trigger buy signals. The author shares results from a July 30 session and plans to fine-tune a LLaMA model on 200+ labeled pattern samples. Code is available on GitHub with a ready-to-use skill on Xiaping Marketplace.

  • •Three-pillar system combines technical, fundamental, and sentiment analysis for Chinese A-shares
  • •Scoring: 30-point system, 12+ triggers buy signal; demonstrated on July 30 trading session
  • •Roadmap includes QLoRA fine-tuning of LLaMA on 200+ labeled samples; open-sourced on GitHub

Generated with AI, which can make mistakes.

#ai-tools#ai-agents#open-source
Read full article at Dev.to

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

The original title is: "MCP Server for Chinese A-Share Quant Analysis in Claude Code and Cursor"
Dev.to
I Built an AI Assistant That Turns Natural-Language Ideas Into Stock Searches
Dev.to
Building an LLM Trading Bot: From Failed Prompts to Visual Chart Inference with Gemini
Dev.to
The original title is "Building a Claude-Powered Stock Screening Pipeline with the EODHD API"
Dev.to