Sublime

Over the last 2 yrs, I've helped 35+ companies improve their AI products. I distilled my approach into this guide, which covers error analysis, synthetic data, eval-human alignment, involving domain experts, optimizing # of experiments & more (1/5) https://t.co/Kn3hyq5zf2

Re: CaseText & eval driven development (EDD) for AI. EDD was always front and center pre-generative AI. Ex: nobody cares about your fraud/churn/forecasting model if it’s not accurate. This is missing with most LLM products. But the best practices and playbooks are there in classic ML!... See more

“PMAX advertisers, welcome to my world” Noa Gutterman dropping a years worth of google testing insights into a 3 minute crash course. So good. - Google UAC only 30% incremental - was able to reduce budget by 70% on Google with no impact to the business https://t.co/drv4Nk5aUy