Need help picking the right emoji (like we did for this post)? 🤔 @cassidoo made an emoji list generator with Copilot CLI. Learn how she did it and pick up tools and tricks for your next project. 👇

If you need more pics: p.s. prize for you on OnlyF💗ns gift free link for Fan💌sly — My contacts: It’s only for my best followers channel ~

$APH Amphenol's CEO Adam Norwitt says inference is growing faster than training and requires an unprecedented intensity of interconnect, and he doesn't need to pick which chip or model wins, because every one of them runs through his product. "We work with customers who make GPU architectures, who make ASIC architectures, who make XPU architectures, you name it." "Because no matter what architecture you're choosing, at the end of the day, to create an AI that really creates value for the end customer, you have to compare everything to everything else. And we see that what that means is you need interconnect to do that." "And so whether it's on training or actually on inference, and to be honest, inference seems to grow at an even faster pace, also requires an intensity of interconnect products that really is unprecedented." "We continue to enable customers in all ways. So we are not model dependent. We're not frontier training versus inference dependent. We're not geographically dependent." "I guess the only thing we are dependent on is that people continue to invest in this thing and it continues to have increases in content. We see that for a long time to come."