The author tries the new Jev action selection model, noting it is unlikely to replace payment fraud detection since many features are tabular and it underperforms LightGBM, yet finds it impressive given this is probably not its intended use case.
Time to start playing with #jev the new action selection model. Will it replace Fraud detection for payment. Well unlikely because a lot of the features are tabular. Underperforming a Lightgbm but it it already impressive given that this probably not the use case for it.
James Long uses Jev to tackle the most annoying problem in personal finance tools: generating good payee descriptions from raw bank data. It got about 95% of the way there on the first try, with many ways to improve the prompting.
A developer uses Jev to decide buy or sell and executes real trades on Kuru's on-chain order book via Monad every 300ms block. Link
Erik Spock Gafni suggests that those who make a lot of profit from Jev trading donate some of their profits to charity.