Trading automation platform

Software package for market analysis with AI-powered indicators and trading agents, which can be changed by the trader and automatically optimized.


Main Features:
  • Smart trading signals, which are based on the analysis of full market information via deep neural networks
  • Smart trading strategies. Particularly, we can optimize such metrics as equity, max drop down, Sharpe Ratio
  • Visual interface for the construction of custom indicators and trading strategies
  • OHLCV + extended trading statistics (Order Book, Order Count Book, Open Interest, Executed Volume, Order Count, and their aggregations by levels)
Performance features:
  • Own data storage format, which doesn't require deserialization of bars and dynamic structures (like orderbook).
  • All software is implemented and optimized in C++.
Other Features:
  • Scalable architecture. It's easy to add new connectors for other exchanges
  • Graphical terminal for manual trading
  • Strategy optimization software
  • API for software developers (Python/C++)
Problem
A lot of traders across the world spend much time on manual trading. Every year this work becomes more and more difficult with increasing amount of automated trading algorithms.
Solution

We used our knowledge in such fields as trading, deep learning, machine learning and high performance computing to develop AI-powered trading environment. Particularly we applied deep learning to analyze full trading order log. For this reason we developed our own data storage and processing algorithms for dealing with 200+ GB of trading information.