img

Things I did in the Bear Market — Spent 2 Years Building A Cryptocurrency Trading Bot Platform

img
valuezone 29 October 2023

Things I did in the Bear Market — Spent 2 Years Building A Cryptocurrency Trading Bot Platform

I would like to start by saying that I’m excited to finally be able to write this article and share this project with the world.


Aesir was built by a small team of developers over the course of two years. This development cycle allowed us to run through a closed and open beta and ensure that Aesir is capable of helping retail investors maximize their gains by offering a platform that’s both easy to use and allows for complex customization of trading strategies.

Aesir originally started as the Binance Volatility bot, an open-source Python script designed to detect and trade volatile assets on the Binance exchange. The application got a lot of attention on Reddit at the time but proved difficult to use as it required some programming knowledge.

In time, Aesir built an identity of its own as we added more features that users wanted. That’s one of the best things about a young start-up, the core team is very close to the user base, and we’re agile enough to constantly incorporate user feedback into the application.

How do Algo Trading Platforms Work?

In order to understand why a tool like Aesir can help a trader find and execute better strategies, it’s good to have a top-level understanding of how they work. Algo Trading Platforms work by allowing the user to create a series of rules, that the platform is going to evaluate and execute. For instance, these rules can be as simple as a weekly DCA:

Buy BTC, ETH, DOT and SOL EVERY 1 Week, With an Order Amount of 100 USDT

The algorithm will then take these parameters and execute trading according to it. This is just a minimal example, you can create strategies that contain complex logic based on Volatility, Technical Indicators and other market tools —

Could also do something like this.

Buy ANY COIN on THIS EXCHANGE that GAINED more than 3% in the 2 MINUTES and the RSI is under 70

It’s this example that truly describes the advantage of trading algorithm over manual trading. A daytrader would simply not be able to analyse over 1000 assets in under a second and make a decision based on this analysis. But here you would be able to catch a coin that’s just about to spike up in price, because you’re constantly evaluating the price of all coins on the exchange.

So in a nutshell, a trading algorithm extends your ability to trade by adding powerful new tools in your arsenal. It goes without saying that the way you use these tools makes a big difference.

The Building Approach

Before actually building the application, we started collecting information on how current algorithmic trading apps such as HummingBot and 3Commas work. We combined this research with the numerous reddit and discord conversations around the kind of features that people expect from a algorithmic trading app, and apart from the usual “I want it to make me profit” we found out that many people tend to value guidance, learning and the ability to grow along with something. Many algorithmic trading platforms out there require the user to already possess specialised knowledge before they can use the application. You can’t really completely remove complexity, since specialised knowledge is needed, but you can make it so people learn from one another, and share the knowledge collectively.

AESIR Features

That’s why we decided to build social mechanics into AESIR. Here is a quick preview of the Leaderboard and Copy Trading —


Leaderboard & Copy Trading

The Leaderboard is a collection of the top performing Configurations on the platform. The ones that are marked as “Public” can be cloned by anyone. Cloning a configuration creates a copy of that algorithmic strategy on your account, and allows you to instantly use it for yourself. The Leaderboard also features gamification mechanics, while encouraging responsible trading. The user with the top performing configuration of the month will receive a special prize as an acknowledgement of their Trading Feats.

Logic Layers

Going back to the example on how algorithmic trading apps work:

Buy ANY COIN on THIS EXCHANGE that GAINED more than 3% in the 2 MINUTES and the RSI is under 70

Once you’ve selected a base and logic layers, AESIR will tell you exactly what the strategy that you created will do:

Test Strategy will trade USDT pairs with an order size of 10 USDT and
a maximum of 3 Open Orders at a time. Assets will not be re-bought
for 10 seconds. Buy and Sell signals for BTC, ETH, TRIBE, BIFI are
checked every 30 seconds and an order will close when Take Profit
reaches +2% or Stop Loss reaches -6%.
1. When a coin pair receives Buy Signals for 2 Moving Average(s) out 
of EMA10, EMA20, SMA20, and 1 Oscillator(s) out of CCI20, AO
AND2. once any of BTC, ETH, TRIBE, BIFI have gained at least 2% in the last 10 minutes an order will be triggered.

Other Features:

  • Trailing Stop Loss (keep an order open while an asset is on the rise)
  • Paper Trading (the ability to test your strategies with fake money)