Development

QuantCoderFS is a comprehensive quantitative finance platform that combines AI-powered equity research and algorithmic strategy generation. Built with FastAPI backend and Next.js frontend, it features two primary modules: Market Intelligence for multi-agent stock analysis with BUY/HOLD/SELL recommendations, and Strategy Generator for automated conversion of academic research papers into QuantConnect-compatible trading algorithms with ~90-95% success rate.

Key Features:

• Multi-agent AI stock analysis (fundamentals, news, price, macro)
• BUY/HOLD/SELL signals with conviction scores (1-10)
• Automated research paper to trading algorithm conversion
• QuantConnect Python code generation (5-phase AI workflow)
• Snapshot-based market analysis tracking
• Integration with multiple AI providers (OpenAI, Anthropic, Ollama)

Note: The GitHub repository contains version 1 of QuantCoderFS, prior to the deployment architecture for free and paid tiers. The current production version includes additional features and infrastructure for multi-tier service delivery.

Chat with Fundamentals is a comprehensive financial research and portfolio management platform that combines AI-powered fundamental analysis with advanced quantitative portfolio optimization. Built as a research engine to explore how autonomous agents and AI can automate and enhance financial analysis, it features multi-agent AI analysis, shares-based portfolio tracking, and RAG systems for quantitative research and SEC filings.

Key Features:

• Multi-agent AI analysis (stocks, ETFs, forex, macro indicators)
• Shares-based portfolio management with 5 optimization strategies (Actual, Equal Weight, MVO, Min Variance, Black-Litterman)
• Comprehensive EODHD API integration (50+ endpoints across 9 categories)
• RAG systems for quant research papers and SEC filings (10-K, 10-Q, 8-K)
• Advanced risk analytics (Monte Carlo, VaR, CVaR, rolling Sharpe ratios)
• Database management for fundamentals, prices, news across multiple asset types
• PostgreSQL + TimescaleDB + Redis + ChromaDB architecture
• Real-time agent console with WebSocket logging

AI-powered document management and query system for maritime conventions including SOLAS, MARPOL, STCW, and other IMO regulations. Uses RAG (Retrieval-Augmented Generation) with vector embeddings to provide accurate, citation-backed answers to regulatory questions. Built with FastAPI backend, React frontend, and Qdrant vector database.

Key Features:

• PDF document indexing with smart chunking (~800 chars)
• Semantic search with OpenAI embeddings (text-embedding-3-small)
• Inline PDF viewer with page navigation and zoom controls
• Chat interface with citation tracking (chapter, section, page)
• Support for 48+ maritime convention PDFs (SOLAS, MARPOL, etc.)

Katja is a comprehensive AI chat platform featuring 5 specialized modes: language learning with grammar correction, general chat, legal reasoning (Saul-7B), medical queries (Meditron/BioMistral), and a terminal-style console for monitoring all interactions. Built with local-first privacy using Ollama, with optional cloud fallback (OpenAI GPT-4o-mini). All data stays on your machine by default.

Key Features:

• 5 specialized chat modes (Language Learning, General, Legal, Medical, Console)
• 7 European languages with real-time grammar correction
• Spaced repetition (SM-2 algorithm) for language learning
• 14+ LLM models including Saul-7B (legal) and Meditron (medical)
• Local-first privacy via Ollama (offline-capable)
• LaTeX math rendering for technical discussions
• Customizable system prompts for each mode
• SQLite database with WAL mode for persistent conversations