Hi,
I am an ex-app developer (McKinsey, WalmartLabs) with a math/statistics background (UC Berkeley, UCLA) originally from Millbrae, California. I use Python, Ruby, Elixir, NodeJS, and learning Rust. Looking for anyone who is interested in developing an automated crypto trading bot on a legitimate exchange like Kraken, Binance or Coinbase. I'm lookng for someone who would want to develop this with me, and/or an investor who can support me to build this. I lost about 90% of my crypto on FTX, so my savings account is kind of starting from scratch, and I'm <i>highly</i> motivated to work now. To avoid what happened to me, we can run the bot on multiple exchanges to distribute counterparty risk and only choose the most trustworthy exchanges like the ones mentioned above. I can work on this 10 hours per day 7 days per week until we make something happen. I just moved to the Philippines in April 2022, so after 5 pm Pacific is the best time for me to talk. That would be my morning.
I was thinking we could backtest a bunch of strategies, then try dry running them, then load up the accounts with real money, small at first, then gradually increase it as our confidence increases in the system. This kind of project would naturally lead to arbitrage opportunities as well. We would split the profits evenly and slowly organically build the trading account. We could even generate fake data that has the same statistical properties as the actual data set, so that we would theoretically have essentially infinite test data -- kind of like video game data. If we wanted to productize this, we can make an interface for users to input and withdraw money from their account. The bot will trade the lump sum using the same algorithm, and we all split the profit equally. If users want to make their own algorithms they can do that too. Ideally you would be an investor/entrepreneur or group of investors, or maybe a developer in a similar position in life, where you can dedicate 70 hrs per week doing this.
There are <i>tons</i> of strategies we can backtest. I'm thinking we could start on spot markets, preferably high time frames, then use neural networks to predict probability of success of a particular trade then use small leverage in proportion to the predicted probability of success. Most trades can have a trailing stop losses to skew the upside potential for every trade.
Anyway, like-minded individuals, please contact me: thirdreplicator@gmail.com
David