

A similar experiment I did comes to mind from 3 years ago.
For the fun of it I was trying to train a few deep neural network configurations (LSTM, a few variations of FCNs, …) to trade shitcoins and downloaded 4 years of 1h candles.
The first easiest idea was to prepare the training data to fire three signals, buy, sell, do nothing (I know a terrible choice). The cost function was setup to do the simple thing and maximize the overall profit (I know an other terrible choice). Fast forward 30min of training and the final outcome is a model that outputs “do nothing” in 100% of the cases.

I don’t even know how to get into that
KDE seemed like quite easy to get started with, I installed their tooling, found a good first issue in Dolphin I could reproduces but couldn’t build dolphin locally following their guide, joined their matrix chat and asked for help, got ignored for a day, then somebody pointed me to another room where I should ask if the build guide was up to date, got nothing their and I gave up
Went through some other interesting open source projects I have been using and I was trying to pick something but to no avail, all issues there is somebody from India that took it and didn’t reply after 4 years (probably they have something in university asking them to do that)
I would rather spend my precious time in building something I have 100% control over than join the rat race. I have cloned a few projects in the past, fixed some annoying issue or upgraded deps to match my environment, installed the forked version and never opened a PR
So yeah, if you have a project and need some extra hand, let me know, I am an expert in Python but interested in languages I am less than an expert in like go, rust, zig, elixir (I am ready to match your style, good quality, skip using LLMs, …) . If interested I could also invite you to help out with a few things I have in progress or I am aiming for a re-write.