Godtrader — Cryptocurrency Research and Trading Management Platform
A private trading desk bringing market monitoring, strategy evaluation and controlled execution into one platform.
What it does
Godtrader is a cryptocurrency research and trading management platform I designed and developed to bring market monitoring, strategy evaluation and controlled trade execution into one place. Integrated with Bybit, it monitors a 55-coin watchlist, records trading opportunities and tracks their outcomes, so decisions rest on measurable performance rather than impression. The platform pairs an interactive dashboard with automated reporting, performance analytics and configurable execution controls.
The problem
Crypto research tends to live across disconnected tools: a charting tab, a spreadsheet of trades, a notes app and the exchange interface itself. Outcomes are rarely recorded in a way that can be reviewed, so it is difficult to tell whether a strategy has an edge or whether a good result was luck. The problem was to build a single system that keeps the market view, the decision record and the execution controls together — with the evidence trail intact.
What made it hard
- Market monitoring, trade records and research notes held in separate tools with no shared source.
- Trade outcomes not recorded consistently, so strategy performance could not be compared over time.
- No deliberate separation between observing an opportunity and acting on it.
- Execution safeguards handled manually, with no guard against duplicate orders or missing stops.
- Reporting assembled by hand rather than generated from the same data the dashboard reads.
Problem analysis
- Separated the platform into three concerns — market intelligence, the trade journal and execution controls — because each answers a different question, and mixing them obscures the evidence.
- Treated the journal as the core of the system: without a reliable record of opportunities and outcomes, performance analysis is anecdote.
- Made live trading disabled by default, so observing the market and executing against it stay deliberately distinct actions.
- Framed the strategy assessment around the limits of the available evidence, rather than presenting a single performance figure as proof.
Solution design
The platform is a Flask application backed by a relational database, with an interactive dashboard in front of it. Bybit supplies market data, and the watchlist defines what is monitored. Every recorded opportunity is stored with its outcome, which lets the analytics layer break performance down by strategy, cryptocurrency and weekday. Execution controls apply position-sizing rules, prevent duplicate orders and check for protective stop-loss orders before anything is sent. Research modules for lending, staking, liquidity mining and initial exchange offerings sit alongside, deliberately unable to move funds. Automated email summaries and PDF reports carry the same record outside the application.
- Market layer
- Bybit integration and a 55-coin watchlist feeding price and market-condition monitoring across multiple timeframes.
- Journal layer
- A relational record of trading opportunities and their outcomes, forming the evidence base for performance analysis.
- Analytics layer
- Strategy, asset and weekday performance breakdowns, risk-adjusted return tracking and a strategy-assessment framework.
- Execution controls
- Position sizing, duplicate-order prevention and protective stop-loss checks — with live trading disabled by default.
- Research modules
- Separate lending, staking, liquidity mining and IEO research surfaces that cannot move funds.
- Reporting layer
- Automated email summaries and PDF reports generated from the same underlying records.
Implementation
- Built the application architecture and backend in Python and Flask, with a relational database behind it.
- Integrated the Bybit exchange API for market data and controlled order handling.
- Developed the dashboard and journal interfaces in JavaScript, HTML and CSS.
- Implemented the strategy-assessment framework so recorded outcomes are examined against the limits of the sample.
- Added execution safeguards — position sizing, duplicate-order prevention and stop-loss checks — with live trading off by default.
- Introduced machine-learning support with LightGBM for signal work, and automated reporting to email and PDF.
- Containerised the application with Docker and deployed it on Dokploy.
Key features
- Monitors a 55-coin watchlist for price patterns and market conditions across multiple timeframes.
- Records trading opportunities and tracks their outcomes in a structured journal.
- Performance analysis by strategy, cryptocurrency and weekday, with risk-adjusted returns.
- Strategy-assessment framework that highlights the limits of the available evidence.
- Execution controls: position sizing, duplicate-order prevention and stop-loss checks.
- Research modules for lending, staking, liquidity mining and IEOs that cannot move funds.
- Automated email summaries and PDF reports.
- Live trading disabled by default and requiring deliberate user action.
Results and impact
- A working platform in active development, integrating live Bybit market data with a recorded trade journal.
- Execution controls and a default-off live-trading switch make the system safe to operate while strategies are evaluated.
- Demonstrates translating a complex financial workflow into practical software with traceable records, clear reporting and deliberate operational controls.