Founded 
Angel Investor. 
Had any background on stock trading before building? Or just went off the info from research papers? 
I built this neural net for high frequency trading as a hopeless side-project. Now it's at almost $10K profit! 
pic.twitter.com/Kru5DBf2aZ
-
-
-
My background in finance was pretty limited. Had a solid understanding of the stock market, but not so much into trading.
-
What really helped were the books listed in this interview (https://www.indiehackers.com/businesses/stock-trading-bot …) + research papers for impl. approaches
End of conversation
New conversation -
-
-
is it just a simulation or you actually tried trading with it
-
has actually been trading intermittently since Sept 2016
-
very cool project and setup :) you sure it didn’t work out because it is a bull market since then?
-
Yes and no. Big percentage of profit comes from liquidity. The stocks used were pretty steady the past months though
End of conversation
New conversation -
-
-
This is really freaking cool! You interested in doing an
@IndieHackers interview? I'm sure lots of ppl would love to read about it -
Sure thing! :)
-
really looking forward to it!
End of conversation
New conversation -
-
-
what types of products/instruments are you trading? I'm interested to know what your latency numbers look like ;)
-
It's mostly heavy volume stocks. I'll post a blog post the following days with more in-depth details :)
End of conversation
New conversation -
-
-
interesting. What was the infrastructure investment worth?
-
Tbh the biggest cost was training the models (Amazon g2.2xlarge), which set me back about $300. Instances were roughly $100/m
End of conversation
New conversation -
-
-
any chance of open sourcing it :D
-
Not sure if I'll open source it, but I'll publish an in-depth blog post on this in a couple days :)
-
fair enough :)
End of conversation
New conversation -
-
-
which programming language ?:)
-
It's C++ and Python on the server. Swift on the GUI-side, sprinkled with a bit of Shell :)
-
interesting setup
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.