Toutes les commandes
Règles
Spécifications
1. Project Overview & Strategic Objective
We are seeking an elite, senior-level MQL developer to design and engineer a bulletproof, proprietary MT4 non-standard bar generation application. The primary objective of this project is to build an independent, institutional-grade charting infrastructure to permanently eliminate third-party plugin dependency risks (highlighted by legacy solutions like AZ-INVEST going unsupported).
The target algorithm currently operates with a high-volume execution profile—generating over 5,000 trades with a 3-hour average holding period and a 33-pip trade expectancy. Because our automated trading systems execute live capital directly off these generated chart structures, absolute data precision, low latency, and structural stability are mandatory.
This contract covers the simultaneous development of an 11-suite non-standard bar application to provide a unified testing and execution platform.
2. Phase 1 Scope of Work: The 11-Bar Suite
The application must generate fully independent, EA-compatible offline charts/custom symbols for the following 11 bar types:
| Bar Type | Technical Description & Requirements | Core Objective |
| 1. Standard Renko | Fixed brick size calculation. Must natively support ATR-based dynamic sizing. | Baseline star performer. |
| 2. Median Renko | Also known as "Better Renko". Smoothed variant requiring wicks to display true historical highs and lows. | Advanced price smoothing. |
| 3. Turbo Renko | Modulates standard Renko characteristics to enable faster signal generation via smaller effective brick structures. | Trend acceleration tracking. |
| 4. Hybrid Renko | Math-driven combination balancing standard Renko block progression with median candle characteristics. | Speed and noise filtering balance. |
| 5. Vanilla Renko | Also known as "PointO". Pure, traditional Japanese charting method containing zero wicks. | Clean rule-based signal isolation. |
| 6. Time-Limited Renko | Standard Renko framework embedded with a hard maximum time-limit parameter per brick. | Prevents frozen charts during low-volatility sessions. |
| 7. Range Bars | Charts built on a fixed price range per bar, capturing true candle wicks within the boundary. | Volatility-based price segmenting. |
| 8. Tick Bars | Bar prints strictly after a user-defined, discrete number of incoming market transactions. | Transaction-activity based charting. |
| 9. Heikin Ashi | Averaged open-high-low-close (OHLC) candles designed to smooth out minor market noise. | Directional trend filter. |
| 10. Kagi Bars | Path-dependent charts that change vertical direction only after a defined price reversal threshold is broken. | Pure price action trend tracking. |
| 11. Point & Figure | Classic, time-independent column charting method mapping price flips based entirely on a defined reversal amount. | Structural support/resistance filtering. |
3. Phase 2 Roadmap: Deferral of Volume-Dependent Bars
Crucial Architectural Note: Volume Bars, Dollar Bars, Delta Bars, and Volumetric Bars have been intentionally excluded from this Phase 1 contract. Because our current deployment is operating in the spot market where decentralized broker tick-volume is unreliable, implementing volume-based bars at this stage is structurally flawed.
We are planning a near-term migration to the CME futures market to trade Micro contracts (MCL and MGC). Once we migrate to real, centralized exchange volume via our Rithmic signal bridge, a separate Phase 2 contract will be issued to the winning developer to build out the volume, dollar, delta, and order-flow volumetric bar types.
4. Mandatory Technical Specifications
The developer must program the application to meet the following structural standards:
· Raw Tick-Data Processing: The generator must construct all offline charts directly from real incoming tick data streams—not M1 bar approximations.
· Full History Engine: Upon initialization or terminal reboot, the application must automatically construct complete historical offline charts stretching back across all available historical tick data.
· ATR-Driven Calibration: Integration of native ATR-based dynamic brick sizing is mandatory for all price-dependent bar variants (Renko, Range, Kagi). The ATR lookback period must be user-definable.
· Real-Time Performance Engine: Multi-threaded, optimized structure capable of pushing real-time chart canvas updates at a minimum frequency of 10 times per second for high-volatility live trading.
· Multi-Instance Support: The architecture must allow multiple simultaneous chart instances of the exact same bar type to run side-by-side with completely different parameters without memory cross-contamination.
· Expert Advisor Thread Safety: Generated charts must pass clean, uncorrupted indicator buffers and OHLC array historical states directly to any standard EA running on the offline chart without latency bottlenecks.
5. Meticulous Acceptance Criteria & Quality Assurance (QA)
Payment will remain held in the MQL5 Escrow system and will only be released upon passing our rigorous multi-stage verification framework:
· The Historical vs. Live Parity Test: We will run the bar generator live for a continuous 48-hour trading period on an asset (e.g., XAUUSD). Concurrently, we will generate the historical chart for that exact same 48-hour period on an identical terminal using raw tick data logs. The two charts must match perfectly down to the single pip. Any evidence of repainting, shifting, or missed bars will result in a immediate revision rejection.
· EA Optimization & Stress Test: We will load a logic-heavy Expert Advisor onto the generated offline charts. We will trigger rapid Walk-Forward optimizations and intensive backtesting simulations. The custom data structure must handle these high-frequency loops without triggering Context Busy, Invalid Price Data, or chart synchronization mismatch errors in the terminal journal.
· Mathematical ATR Verification: Under high-volatility conditions, we will cross-verify that the automated ATR-sized brick parameters dynamically adjust to match the exact mathematical output of the underlying timeframe's ATR over the designated lookback period.
· Resource Profiling & Leak Checks: The generator will be profiled across 10 concurrent active offline charts. The application must show highly efficient memory usage, zero memory leaks, and stable CPU consumption without freezing the MetaTrader terminal.
6. Developer Requirements & How to Apply
We are only entertaining bids from elite, systematic developers who treat code as high-performance financial infrastructure.
· Platform Credentials: You must hold a minimum 4.5-star rating on the MQL5 Freelance Marketplace and have successfully completed at least 20+ verified projects.
· Domain Expertise: You must possess verifiable, deep experience in engineering custom symbol structures, handling offline chart generation (post message threading), and managing raw tick data engines within MetaTrader 4.
· Application Rule (Code Samples Mandatory): When submitting your proposal, you must provide sanitized code snippets, documentation examples, or structural overviews of prior non-standard bar engines you have authored. Generic, automated bids that fail to attach proof of past offline charting experience will be ignored.
· NDA Requirement: The selected developer will be required to execute a strict Non-Disclosure Agreement (NDA) prior to the handover of any proprietary strategy criteria or existing EA frameworks for compatibility testing.
Please provide a comprehensive, transparent quote for the entire 11-suite development cycle, along with your projected chronological timeline for milestones and alpha delivery.