Skip to content
Anas KhanLet's talk

TradeM8

Chart patterns meet market context.

Project
AI-powered crypto trading assistant
Year
2025–2026
Discuss this project
Trading workspaceMulti-timeframe charting, indicators, pattern detection and the asset watchlist in one workspace.
Detected chart patternA captured ETHUSDT frame with the detected double-top region annotated by the vision workflow.
ETH market analysisA generated analysis combining the recommendation, rationale and supporting values. Figures belong to this captured session.
Market overviewThe admin view for reviewing market coverage, article sentiment, signals and account activity.
Sentiment pipelineRuntime output from the scheduled news-ingestion and sentiment-analysis pipeline.
Analysis serviceThe analysis service receiving chart data and passing the annotated frame through the model pipeline.
NEAR market analysisA second generated analysis showing how the interface handles a HOLD recommendation. Figures belong to this captured session.
TradeM8
TradeM8 trading workspace with a candlestick chart, chart controls and an asset watchlist
Trading workspace

Multi-timeframe charting, indicators, pattern detection and the asset watchlist in one workspace.

1 / 7

The problem

A chart pattern describes only one part of a market. TradeM8 brings visual pattern detection, technical indicators, financial news, and social sentiment into a confidence-based analysis experience.

Implementation

The platform uses Python and FastAPI with a React interface and PostgreSQL. YOLOv8 and Transformers form part of the AI stack. CryptoPanic, RSS, and CoinGecko pipelines supply market context alongside chart-pattern analysis.

Workflow overview
  1. Charts and market feeds
  2. Patterns and sentiment
  3. Indicator validation
  4. Confidence-based analysis

Features

  • Chart-pattern detection

    Analyze visual chart patterns with a computer-vision workflow built around YOLOv8.

  • News and sentiment analysis

    Bring financial news and social sentiment into the same analysis as chart patterns.

  • Technical validation

    Evaluate detected signals with EMA, RSI and ADX for additional technical context.

  • Connected market feeds

    Combine CryptoPanic, RSS and CoinGecko pipelines for news and market context.

Outcomes

validation indicators
3
EMA, RSI and ADX
  • Validated signals with EMA, RSI, and ADX technical indicators.
  • Connected CryptoPanic, RSS, and CoinGecko data pipelines for market context.

The engineering focus is the integration of computer vision, language-based analysis, external data pipelines, and a full-stack interface for reviewing their combined output.

System illustration showing CMS content passing through page generation into a public website
Next projectTecaudex Website & CMS