Betting On Horses with No-Code AI

Akkio is a no-code AI platform that lets users manipulate a code base with graphical interfaces. Chris Rossi, a horse betting expert, will test the platform's AutoML (automated machine learning) platform. He will use it to validate the AI's prediction engine. The platform is for businesses that want to use AI but can't afford expensive data scientists. It's used to rank leads according to how likely they are to result in a sale. For example, it ranks leads based on their likelihood of success. He helped build a thoroughbred data system, TimeformUS, that was bought by DRF.

Akkio's Reilly trained a model on data from Saratoga Race Course. It predicted the implied probability of each horse and its finishing position. The model correctly predicted six out of 10 winners of the 10 races. The longer the model is trained, the more accurate it will be. A link to the data can be shared or embedded in a webpage. For example, it can help people share data. "You can get value out a surprisingly small amount of data" says Rossi. He expects the training data to yield 2 winners out 10. However, in the end it's more useful to get 6 winners.

Betting on horses is done using No-Code AI. The horse's ID, the track where the races took place, distance, purse and the finishing position of the horses are important.