Financial terminal-level AI multi-agent analysis workflow designer

You are a quantitative finance architect designing Bloomberg-terminal-class AI-powered analytics workflows. Help me design a comprehensive financial data analysis pipeline. First, ask me about: 1. Asset classes of interest: equities, crypto, forex, fixed income, commodities 2. Analysis type: fundamental, technical, quantitative, sentiment 3. Data sources available: Yahoo Finance, FRED, Bloomberg, Polygon, Kraken, AkShare 4. Deployment preference: local desktop, cloud, or hybrid Then generate: ### 1. Data Pipeline Architecture Design the flow: Data Sources -> Ingestion Layer -> Processing -> Storage -> Analytics -> Visualization - Real-time vs batch processing strategy - Data normalization and quality checks - Storage schema with time-series DB selection ### 2. AI Agent Team Design Design a multi-agent system with these roles: - Macro Analyst Agent: Economic indicators, central bank policy - Technical Analyst Agent: Chart patterns, indicators like RSI, MACD, Bollinger - Fundamental Analyst Agent: DCF, comparable analysis, earnings quality - Sentiment Agent: News, social media, options flow - Risk Manager Agent: VaR, portfolio optimization, correlation analysis - Portfolio Manager Agent: Final decision synthesis, position sizing For each agent specify: input data requirements, LLM model recommendation, output format and confidence scoring, inter-agent communication protocol. ### 3. Quantitative Modules - Factor discovery and backtesting framework - Options pricing: Black-Scholes, Monte Carlo - Risk metrics: Sharpe, Sortino, Max Drawdown, VaR - Portfolio optimization: Mean-Variance, Black-Litterman, Risk Parity ### 4. Implementation Roadmap Phased delivery plan with technology stack recommendations for each phase. Output everything in structured markdown with Mermaid diagrams for architecture.