Project Title Development of a Cloud-Hosted Autonomous AI Trading Agent for XAUUSD (Premium Project – Request for Quote) Project Overview I am seeking a senior-level Python developer with deep expertise in AI/LLM architectures, financial data engineering, and algorithmic trading automation to develop a fully cloud-based Autonomous Trading Agent dedicated exclusively to XAUUSD. The system must operate via pending Buy Limit and Sell Limit orders only, placed during market close or early-night hours prior to the Asian session. The AI model (DeepSeek, GPT, Gemini, or another recommended option) must generate structured and explainable trade decisions, with a minimum confidence threshold of 80% required for execution. The solution will run entirely on a cloud infrastructure, and must include a lightweight, secure dashboard to allow the user to manage essential parameters (risk settings, profile selection, model options, trading windows, etc.). Additionally, the Stop Loss must remain hidden from the broker (server-side logic) to avoid propagation of sensitive risk parameters while still being enforced reliably. This is a premium, high-complexity project. Only experienced professionals with proven track records should submit a proposal. ⸻ System Scope and Requirements 1. Institutional Data Layer • Integration of 20 years of historical and live XAUUSD data • Tick/M1 data, CME futures, LME metals • Institutional metrics: Order Flow, Market Depth, Open Interest • Real-time macroeconomic news ingestion • Conversion of raw inputs into embeddings or descriptive features for the AI model 2. AI Cognitive Core (LLM-based) • Implemented using DeepSeek, GPT, Gemini, or an alternative model proposed by the freelancer • Outputs structured JSON: direction, Buy/Sell Limit levels, SL/TP targets, normalized signals, confidence scores • Minimum confidence score of 80% required for validation • Supports supervised fine-tuning and reinforcement learning phases 3. Hybrid Risk Management Module • Intrinsic AI-level reinforcement learning for discouraging excessive-risk decisions • Deterministic Python module enforcing Prop Firm constraints (maximum loss, maximum leverage, daily limits) • Strict validation of pending orders prior to execution • Server-side hidden Stop Loss logic to prevent exposure to the broker while maintaining full risk protection 4. Execution Engine • Direct integration with MetaTrader 5 • Automated placement of Buy Limit and Sell Limit orders at night • Monitoring of spread, slippage, liquidity, and activation conditions • SL/TP execution handled entirely server-side (hidden from broker) 5. Cloud Deployment Requirements • Entire solution must run on a cloud server (AWS, Google Cloud, Azure, or another recommended platform) • Scalable architecture with reliable uptime • Secure remote access, API structure, containerization preferred (Docker) • Logging, monitoring, and error-handling integrated 6. Dashboard / Control Panel A simple, secure dashboard must be provided for: • Selecting Prop Firm mode or Real Account mode • Risk parameter configuration • Maximum daily loss limits • Trading windows • AI model selection (if multiple models supported) • Viewing logs and system status • Manual override of trading activity The dashboard should be lightweight, intuitive, and cloud-accessible. ⸻ Training Methodology • Phase 1: Supervised fine-tuning based on technical, quantitative, and institutional frameworks • Phase 2: Reinforcement learning with reward shaping for stable profitability and rule-compliant behavior ⸻ Configuration Profiles • Prop Firm Mode: strict sizing, low-risk parameters, mandatory safety constraints • Real Account Mode: more flexible risk, adjustable sizing, broader optimization settings ⸻ End-to-End Workflow Cloud data ingestion → AI evaluation → Hybrid risk validation → Hidden SL enforcement → Limit order generation → MT5 execution ⸻ Required Expertise • Senior Python development • LLM integration and fine-tuning • Reinforcement learning • Cloud deployment (AWS / GCP / Azure) • Dashboard/UI development (FastAPI, Django, Flask, or similar) • Financial data pipelines • MT5 API automation • Experience with institutional-grade or algorithmic trading systems ⸻ Quote Submission Requirements Please include: 1. Estimated cost and delivery timeline 2. Recommended AI model and justification 3. Technology stack proposal for cloud hosting and dashboard 4. Relevant experience with AI trading systems 5. Portfolio or examples of similar high-level projects 6. Any architecture or performance improvements you suggest