Senior MQL5 Expert: Risk-Free Execution Logic (40/30/30)

Tarea técnica

This is the Requirements Specification for your project, formatted to be clear, concise, and structured. This document will allow a professional MQL5/Python developer to accurately estimate the complexity, time, and cost of building your IATS.
IATS Project: Expert Advisor (MQL5) & Python Hub Specification
Job Title: AI-Validated Algorithmic Execution System Developer (MQL5/Python)
Project Goal: Build the high-speed execution engine and orchestration layer for a risk-first automated trading platform.
I. System Architecture & Environment
| Component | Responsibility | Technology | Requirements |
|---|---|---|---|
| Execution Engine | Trade Management, Data Feed, Auditing. | MQL5 (MT5) | Must be compiled for MetaTrader 5 (64-bit). |
| Orchestration Hub | AI Calls, Risk Calculation, Dashboard Server. | Python 3.10+ (with requests, Streamlit, SQLite) | Must run reliably 24/7 on a Windows VPS. |
| Communication | Bidirectional link between Python and MQL5. | Local HTTP/WebSockets (or file-based bridge) | Must ensure near-instantaneous (low-latency) command execution. |
| Risk Boundary | Max Risk Per Trade. | 1.0% of current account equity. | Non-negotiable hard limit. |
II. Module 1: MQL5 Expert Advisor (The Engine)
The MQL5 EA must perform three core functions: Execution, Management, and Data Fetching.
A. Execution and Lot Sizing
 * Instruction Input: The EA must constantly listen for and accept a single JSON command payload from the Python Hub, containing: Symbol, Type, Entry, SL, TP1, TP2, TP3, and the three pre-calculated Lot Sizes (\mathbf{40\%/30\%/30\%}).
 * Order Placement: Upon receiving a command, the EA must immediately open three separate market orders (Orders A, B, C) simultaneously for the target symbol, ensuring all orders share the same Initial Stop Loss (SL) but have unique Take Profit (TP) levels (TP1, TP2, TP3, respectively).
 * Unique Tracking: All three orders must be placed with a shared, custom Magic Number for unit tracking.
B. Mandatory Trade Manager (40%/30%/30% Protocol)
This logic must be implemented inside the OnTick() loop and use status flags to ensure rules are only executed once per trigger.
| Rule | Trigger Condition (Example: BUY Trade) | Required Action (Using PositionClosePartial and PositionModify) |
|---|---|---|
| Rule 1: Risk Elimination | Current Price (Bid) >= TP1 | 1. Close Order A (40% volume). 2. IMMEDIATELY Modify SL for the remaining \mathbf{60\%} (Orders B & C) to the Initial Entry Price (Breakeven). |
| Rule 2: Profit Securing | Current Price (Bid) >= TP2 and Rule 1 is complete. | Close Order B (30% volume). |
| Rule 3: Final Close | Current Price (Bid) >= TP3 and Rule 2 is complete. | Close Order C (30% volume), concluding the trade cycle. |
C. Data Feed and Auditing
 * Market Context Feed: The EA must calculate and package two pieces of data to send to the Python Hub upon request:
   * Trend Bias: Current price position relative to the 200-period EMA on the H4 timeframe (ABOVE or BELOW).
   * Volatility Check: Current Bid/Ask Spread in pips.
 * Audit Data Capture: Upon complete closure of the trade unit (Orders A, B, C are closed), the EA must calculate and log:
   * Final P/L (Profit/Loss).
   * Trade Duration.
   * Maximum Intrade Drawdown (MID): The largest unfavorable movement suffered by the position from the entry price before final closure.
III. Module 3: Python Orchestration Hub
The Python Hub manages external communication, the AI pipeline, and the final risk gate.
A. Core Risk and Lot Sizing
 * Maximum Risk Enforcement: The Hub must calculate the Total Lot Size based strictly on the current account balance and the \mathbf{1.0\%} risk limit, using the \text{SL} distance extracted by Agent 1.
 * Volume Allocation: The calculated Total Lot Size must be split into 40\%, 30\%, and 30\% components before being sent to the MQL5 EA.
B. Three-Stage Gemini AI Pipeline
The Hub must execute three sequential, independent calls to the Gemini 2.5 Flash API (TEXT_API_URL) for every incoming signal. The pipeline halts on any Veto signal.
| Agent | Input Data | Output Veto Rule |
|---|---|---|
| Agent 1: Translator | Raw Telegram Text | VETO: If JSON output is malformed or critical data (SL, TP) is missing. |
| Agent 2: Auditor | Raw Telegram Text | VETO: If Confidence Score is below 80 (threshold must be dynamically adjustable via the Streamlit Dashboard). |
| Agent 3: Strategist | Clean JSON from Agent 1 + Market Context Data from MQL5 | VETO: If the AI determines the trade violates a strategic rule (e.g., trading against the H4 200 EMA, or Spread > 2.0 pips). |
C. Dashboard and State Management
 * Streamlit Integration: Host the Streamlit web application on the VPS, dynamically reading and writing configuration variables (the three AI Prompts and Veto thresholds) to the SQLite database.
 * Live Data Integration: Pull and display current equity, P/L, and the logged MID/Audit data from the SQLite database onto the dashboard.
 * Gemini Dashboard Features: Implement the Auto-Refine Prompt (Text Generation API) and the TTS Weekly Recap (TTS API).
IV. Required Expertise & Timeframe
| Requirement | Level | Note |
|---|---|---|
| MQL5 Development | Expert | Deep experience required with complex order modification (PositionModify) and partial closures (PositionClosePartial). |
| Python Development | Expert | Required for multi-API orchestration (Gemini, Telegram), database management, and web framework (Streamlit) deployment. |
| Development Time | 8 Weeks | This is the minimum required timeline for clean coding and rigorous cross-platform testing of the complex bi-directional communication layer and trade management logic. |

Han respondido

1
Desarrollador 1
Evaluación
(22)
Proyectos
30
20%
Arbitraje
8
63% / 13%
Caducado
9
30%
Libre
2
Desarrollador 2
Evaluación
(12)
Proyectos
47
85%
Arbitraje
1
0% / 100%
Caducado
9
19%
Libre
3
Desarrollador 3
Evaluación
(325)
Proyectos
505
19%
Arbitraje
32
44% / 31%
Caducado
34
7%
Trabajando
4
Desarrollador 4
Evaluación
(168)
Proyectos
199
48%
Arbitraje
3
33% / 33%
Caducado
2
1%
Libre
5
Desarrollador 5
Evaluación
(2)
Proyectos
1
0%
Arbitraje
1
0% / 100%
Caducado
0
Libre
6
Desarrollador 6
Evaluación
(13)
Proyectos
13
38%
Arbitraje
1
0% / 100%
Caducado
1
8%
Libre
7
Desarrollador 7
Evaluación
(3)
Proyectos
3
33%
Arbitraje
0
Caducado
0
Libre
Ha publicado: 2 artículos
8
Desarrollador 8
Evaluación
(73)
Proyectos
83
64%
Arbitraje
5
100% / 0%
Caducado
9
11%
Libre
9
Desarrollador 9
Evaluación
(16)
Proyectos
35
23%
Arbitraje
4
0% / 50%
Caducado
2
6%
Trabaja
10
Desarrollador 10
Evaluación
(295)
Proyectos
473
40%
Arbitraje
103
41% / 23%
Caducado
78
16%
Ocupado
Ha publicado: 2 ejemplos
11
Desarrollador 11
Evaluación
(16)
Proyectos
20
0%
Arbitraje
10
0% / 80%
Caducado
6
30%
Libre
12
Desarrollador 12
Evaluación
(12)
Proyectos
16
13%
Arbitraje
4
50% / 25%
Caducado
4
25%
Trabaja
13
Desarrollador 13
Evaluación
Proyectos
1
0%
Arbitraje
1
100% / 0%
Caducado
0
Libre
Solicitudes similares
Phahla fx boto 30+ USD
99.99% signal accuracy 10-15 trades distribution all currency trade and meta AI assistance on loss[advice] stop and start robot cyber security firewall protection activation code: 20060605TLP20 Please create a trading bot with any logo with the name elevation
Refine signal trigger execution . Optimize live chart performance . Ensure stable and clean code structure : Stable and clean code is important . Otherwise its a mess . Apply with as much accurate structure you foresee
Here is what I’m looking for: • Signal source: On-chain wallet address monitoring. I want to copy trade high win-rate meme coin wallets (smart money / profitable traders), not Telegram signals. • Market: Crypto (primarily meme coins). • Execution speed: Very important — I want as close to sub-second execution as possible. Low latency is a top priority. • Infrastructure: I’m open to self-hosted VPS or cloud setup if
This post is subject to developers response . Edit the post as you like . May be with me you can make a come back . So , , , Shift calculations . More to the calculation then you can comprehend is known . What else comes to your mind
I’m hiring an experienced MQL5 developer to finish and fix an existing project (NOT building from scratch). I have: An existing MT5 EA (.mq5 + .ex5) that is based on my TradingView logic A TradingView indicator version used for signals/alerts The EA works but has logic/consistency issues and needs improvements + cleanup Goal Make the EA reliable and consistent: Ensure entries/exits match the intended logic Fix
The indicator a bit inverted. But it doesn’t matter to me as long as the winrate make sense for investment. For brief details regarding the indicator. What should have been a sell, i inverted it into buy with sl and tp swapped(only change the name of sl and tp for visualisation , but the code still on right tp and sl) . And in script ive inverted the signal command code. But the trouble is the tp and sl cant be
Ai robot 30 - 50 USD
1️⃣ System Architecture An AI robot typically consists of the following subsystems: 🔹 1. Perception Layer Collects environmental data using: RGB / Depth cameras LiDAR Ultrasonic sensors IMUs (Inertial Measurement Units) Microphones Data is processed using: Computer Vision (e.g., object detection, SLAM) Signal processing Sensor fusion algorithms 🔹 2. Cognition / Intelligence Layer Implements AI models such as
Trailing Stop Based on Thresholds . Other Necessary Filters already Coded . Live Chart Only . The strategy already coded - needs a fresh new draft . To Start from Signal Trigger
What informtion would you need for Ninjatrader futures automated trading and how long would it take ? if anyone can give me answer i will be happy to discuss more about the project thanks fill free to bid to the project thanks
I’m seeking an experienced developer to create an AI-powered trading system integrated with MetaTrader 5. The scope includes delivering a fully functional MT5 module, an AI component a compatible language, and complete source code for both. This will focus exclusively on the core trading logic and AI integration UI elements or dashboards are not required. Apply only if you have proven experience with AI integrations

Información sobre el proyecto

Presupuesto
100+ USD
Plazo límite de ejecución
de 1 a 20 día(s)