← Back to portfolio

Algorithmic Trading Bot

A Python project designed to create a trading bot that mimics an actual strategy used by a real trader.

Python Trading Automation

At a glance

Use this section to quickly navigate to key topics on this page.

Contents

Jump to section

Role

Solo builder and owner. I am designing the strategy experimentation workflow and implementing the core bot pipeline with a safety-first approach, including paper testing before any live use. The trader provided the strategy and detailed approach the bot was to take.

Stack: Python

Scope of work

This project aims to build a bot that mimics a trader's approach to trading, following a predetermined strategy: market reading and setting parameters of the trade (entry, exit, and take profit).

  • Market Reading: a clear structure for reading charts in the way a trader would read them, in the context of the strategies.
  • Strategy Understanding: ensuring the bot really understands the strategy and how it applies after reading the chart, checking if the chart lines up with a potential trade for the strategy.
  • Backtesting and Paper Trading: test modes to ensure the bot is working properly, with no live trades executed.
  • Risk controls: position sizing, limits, and validation checks are integrated into the workflow.

How it works

  • Analyze: pull market data (historical + live).
  • Check: verify that the market data matches the potential readings for the strategy.
  • Generate signals: calculate indicators and conditions to produce trade intents.
  • Simulate execution: model orders, fills, fees, and slippage assumptions. Only simulation on test mode; in real mode it would execute here.
  • Evaluate: summarize trade information: entry points, take profit points, stop loss points and report back once the trade hits take profit or stop loss.

Figures

Diagrams follow the same steps as How it works and Scope of work. The screenshot is real output from a signal through execution.

Pipeline: Analyze, Check, Generate signals, Simulate execution, Evaluate with entry SL TP
Pipeline: Analyze (market data) → Check (strategy fit) → Generate signals → Simulate execution (test mode) or live → Evaluate (entry, stop loss, take profit, reporting).
Scope alignment: market and strategy, paper backtest, risk controls, gated live
Scope on one screen: market reading + strategy understanding happen before signals; paper/backtest with no live trades in test mode; risk sizing and validation in the workflow; live only after proof.
Discord bot messages: trade signal with pair, direction, entry, stop loss, target and stake, plus execution confirmation with trade ID
Example: alert with entry, stop, target, and stake, then execution confirmation with ID and timestamp (matches Evaluate / reporting).

Results

  • Clear pipeline: a structured process from reading the market to executing trades.
  • Safety-first defaults: paper and backtest workflows ensure the bot is actually able to read the strategy (strategy is tested by the trader).
  • Strategy match: the bot replicated the trader’s approach, taking the same trades they would have, at higher frequency and across multiple forex pairs.

Future features

  • Strategy library: plug-in strategy modules with standardized inputs/outputs.
  • Better execution modelling: improved fee/slippage assumptions and order types.
  • Live mode guardrails: hard risk limits, circuit breakers, and monitoring before any real capital.