Richard Ho, VP of Hardware at OpenAI, explains how the team built Jalape帽o, its first custom AI accelerator, and why controlling more of the hardware stack could lower inference costs and improve performance.聽

Subscribe to the Gradient Flow Newsletter 馃摡聽 https://gradientflow.substack.com/

Subscribe: Apple 路 Spotify 路 Overcast 路 Pocket Casts 路 AntennaPod 路 Podcast Addict 路 Amazon 路聽 RSS.

Detailed show notes and transcript, can be found on The Data Exchange web site.

Rss Apple Podcaster
Podden och tillh枚rande omslagsbild p氓 den h盲r sidan tillh枚r Ben Lorica. Inneh氓llet i podden 盲r skapat av Ben Lorica och inte av, eller tillsammans med, Poddtoppen.