The Biggest Sin in AI Inference Right Now | Deep Geeks ep07

The SaaS world is arguing about whether AI features have product-market fit and what to charge for them. Val Bercovici says that’s half the conversation. The other half is cost: Most inference today skips cache reads and repeats redundant prefills, which he calls the biggest sin in inference right now, and almost every provider commits it. https://www.weka.io/podcast/ai-broke-the-yardstick-now-what Learn more: - Listen to the full episode: https://www.weka.io/podcast/ai-broke-the-yardstick-now-what - WEKA careers: https://www.weka.io/careers/?utm_source=youtube&utm_medium=social&utm_content=careers - Subscribe for more AI infrastructure content: https://www.youtube.com/@WekaIO?sub_confirmation=1