I’m not a dev, nor do I have any experience coding, but given that the sx bet api is open and free to use, I wanted to find out if it's possible to code a working exchange betting bot entirely with AI. Turns out it's not very hard with the right approach.
I’ve since built a bot that I use to help me post and dynamically update maker offers across numerous markets simultaneously, which was completely impossible for me to do manually before. I’m historically a pretty bad bettor, but layering in some automation has allowed me to 1) get 2-5% better odds on every single pick 2) never leave a stale offer up.
Through a good amount of trial and error, I found that the recipe to coding just about any programmatic betting bot/tool with AI is to have super simple standalone modules showing how to interact with the exchange API (e.g., fetching current odds for a market or placing a bet), + a clear, plain-english description of the actions the bot needs to make for your betting strategy (could be hitting +EV, arbing, hedging, scalping, market making etc.) and the AI can implement it surprisingly well.
If you want to see the code for a bot built with this strategy (+ the prompt and supporting context I used), I’ve uploaded it all to Github: https://github.com/declansx/odds-maximizer-betting-bot
I’ve since built a bot that I use to help me post and dynamically update maker offers across numerous markets simultaneously, which was completely impossible for me to do manually before. I’m historically a pretty bad bettor, but layering in some automation has allowed me to 1) get 2-5% better odds on every single pick 2) never leave a stale offer up.
Through a good amount of trial and error, I found that the recipe to coding just about any programmatic betting bot/tool with AI is to have super simple standalone modules showing how to interact with the exchange API (e.g., fetching current odds for a market or placing a bet), + a clear, plain-english description of the actions the bot needs to make for your betting strategy (could be hitting +EV, arbing, hedging, scalping, market making etc.) and the AI can implement it surprisingly well.
If you want to see the code for a bot built with this strategy (+ the prompt and supporting context I used), I’ve uploaded it all to Github: https://github.com/declansx/odds-maximizer-betting-bot