Our client needed multiple automated strategies to be coded specifically for trading in crypto markets. The challenge our team was presented with, was not only to program the trading strategies but to also integrate rich historical market data for detailed simulated trading performance reporting, that included tick by tick bid/ask data.
Considering the challenge we determined a custom data feed could be built for Tardis.dev to access historical tick and bid/ask data, and with that feed we could build an adapter to feed into QuantConnect's LEAN engine where we would build the custom strategies.
Client: Proprietary Trading Firm
Our team set up a DigitalOcean server hosting QuantConnect’s LEAN engine running locally. Then we wrote the C# code on QuantConnect for all of the automated strategies, with specific requirements for crypto markets.
From there we custom built a data adapter using Python on the local machine for feeding the Tardis data via CSV into the LEAN engine. We then built a custom UI with Blazor Boilerplate connecting to the strategies we hosted on the LEAN engine for on-demand strategy simulation and reporting utilizing historical and real-time data for any coin-pair hosted on Tardis.