Hi HN,I built QuantDinger: an open-source, privacy-first quantitative trading platform that runs completely locally (or self-hosted).It covers the full trading lifecycle in one Python-native system:AI-assisted strategy creation (visual Python editor + LLM co-pilot) Powerful backtesting & optimization (vectorized, multi-threaded) Live execution across markets: crypto (10+ exchanges via CCXT), US/HK stocks (Interactive Brokers), forex (MT5) Multi-agent LLM research system with local RAG and reflection loops Dashboard with real-time charts, portfolio monitoring, global market overview

Key differentiators:Local-first & privacy: Your data and API keys never leave your machine/server. No cloud vendor lock-in. AI integration: Supports multiple LLMs (OpenAI, Gemini, Grok, DeepSeek, OpenRouter) for strategy generation, market sentiment, parameter tuning, etc. Multi-language UI: English, Chinese, Japanese, Korean Docker Compose one-command deploy

GitHub: https://github.com/brokermr810/QuantDinger Live demo (hosted version): https://ai.quantdinger.com Intro video: https://youtu.be/HPTVpqL7knMCurrently at ~440 stars. Recently added global market monitoring, fixed position sync bugs, better mobile profile page, and expanded LLM support.I'm solo-developing this (with some community help) and would love feedback, especially on:Strategy examples or use cases you find useful Pain points in current quant tools that this could address Bugs / feature requests (issues welcome) If anyone wants to contribute code/docs/translations

Thanks for looking — happy to answer questions!(Apache 2.0 license)