Hi, great question @itsanas. And I agree–those terms can be overused.
We do, in fact, use machine learning. Primarily, we’re using Azure’s ML features. But we also use BigML and certain features on AWS. Last, COIN-OR has some really nice optimization toolsets that help us tweak and adapt the models.
Here are some posts you might like:
An earlier discussion: https://crypto-ml.com/how-to-use-machine-learning-to-trade-bitcoin-and-crypto/
Market Index: https://crypto-ml.com/how-the-market-index-works/
Apart from that, we have quite a few articles on our blog that go into some of the various aspects and features of our models: https://crypto-ml.com/blog/
Thanks and let me know if I can help answer anything else.