vibe-investing  by gameworkerkim

AI-powered financial analysis and trading ecosystem

Created 2 months ago
272 stars

Top 94.6% on SourcePulse

GitHubView on GitHub
Project Summary

Summary

Addresses market noise and alpha generation using AI. Offers LLM-powered quant trading tools, backtesting, and curated resources for NASDAQ, S&P500, and crypto. Targets engineers and power users seeking data-driven insights, benefiting from a multi-LLM committee approach for robust, verifiable investment decisions.

How It Works

Employs an "agentic pipeline" where LLMs invoke tools to analyze market data. A core "multi-LLM committee" cross-validates insights from diverse models (Claude, Gemini, ChatGPT, DeepSeek) to simulate a robust investment committee, mitigating single-model biases. Emphasizes human oversight and verifiable reasoning trails, positioning AI as a co-pilot.

Quick Start & Requirements

Installation varies: npm install (Node.js), pip/python (Python), npx (CLI), or git clone/npm install (web apps). Prerequisites include Python, Node.js, npm, PostgreSQL, TimescaleDB (optional: Ollama for offline LLMs). Some tools offer MOCK modes or require no API keys. Direct execution commands provided for key tools.

Highlighted Details

  • Features "Awesome" curated lists evaluating 50+ tools/prompts with scoring and common pitfalls.
  • Includes novel tools like ARDS-X (real-time regime classification, no API key) and LAON VaultGuard (offline, multi-LLM Git secret scanning).
  • Highlights multi-LLM "committee" for cross-validation and provides detailed backtesting results (e.g., AMQS US: CAGR 38.75%, MDD -16.9%).

Maintenance & Community

README updates are infrequent. Contributions welcomed via GitHub stars, issues, and PRs. Author: Dennis Kim (Betalabs Inc. CEO, ex-Microsoft Azure MVP). No explicit community channels or roadmap links provided.

Licensing & Compatibility

MIT License permits free use, modification, and distribution with attribution. Suitable for commercial use, requiring attribution for columns and papers.

Limitations & Caveats

LLMs are tools requiring human insight and verification; generated code may contain errors/biases. Backtesting results are idealized. Project is for research/education only, with no profit guarantees. Managing others' funds requires licensing. High-risk instruments discussed; user responsibility is paramount.

Health Check
Last Commit

21 hours ago

Responsiveness

Inactive

Pull Requests (30d)
1
Issues (30d)
0
Star History
97 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.