← sources

A DeepSeek engineer just said the thing I've been feeling about AI for months

source: reddit (reddit:artificial)published 2026-09-16 16:59ingested 2026-09-17 02:35

retained text

I just read the essay by a guy at DeepSeek who wrote the attention kernel for their latest model, and it hit me in a weird spot. He basically says he knows AI will do his job better than him within a year. He's not mad about it he's not scared he's just sad about the quiet afternoons, the ones where he'd sit and write operators, it felt like a game he was leveling up in, I feel the exact same way. And he's going to keep doing it anyway because it's fun, and because if he stops, someone else doesn't. He'd rather be the one doing the replacing than the one getting replaced.

He's still at his desk right now, still writing the code, and he already knows what's coming and he's not running from it. He's just finishing the thing he loves while it still matters.

Anyway, I don't think about this as China vs USA. I think about it like the gaming market. Every year someone ships something better, and yeah we're getting close to saturation, but that doesn't make any one year less cool. I don't only play American games, I don't only drive American cars. I care about the thing, not the flag.

What I want is for the race to keep going. DeepSeek ships something, OpenAI ships something, someone in Korea or Brazil or wherever ships something. I want the arms race. I want every year to be the year where I'm like "oh wow, that actually works."

The only version that actually scares me isn't "China wins" or "USA wins." It's one company wins and locks the door. That's the cyberpunk thing. That's the one that makes me sad.

So when he says he stayed at DeepSeek because they're open source, that's not him picking a national team. He just wants the door to stay open. He wants the next person to build on what he built. That's the whole thing. Not geography, not flags, just whether the next person gets in the room or stands outside the glass.

I don't know if AI saves us or hurts us. I don't know if his students are going to be fine or if they're going to ship a million broken systems with AI doing the thinking for them. I don't know.

But I know I'm on the side of the door staying open. And I bet most people are too if you just ask them straight. We don't usually get asked though.

document fetched 2026-09-17 00:19 · canonical https://reddit.com/r/artificial/comments/1wi2zm9/a_deepseek_engineer_just_said_the_thing_ive_been