Nitesh Seram

Engineer @greatfrontend

[Nitesh Seram @niteshseram]

@imthepk https://t.co/hTNyxVheha

[Nitesh Seram @niteshseram]

checking twitter to see which new AI model dropped while i was busy using the last one

[Nitesh Seram @niteshseram]

can we keep one thing for more than a week please.
i haven’t even processed Jev yet

[雨夹雪❄️ @mizorewww]

介绍比Jev快50倍，在你设备上跑的laya-mlx!

只在你的设备上占用最高1G内存

Laya是一个开源的类似于Jev的,基于文本输出概率的分类系统

我将其移植到MLX,并且做了一些性能优化!

视频中就是这个模型在我的本地M3Max上玩贪吃蛇

这个模型能够以每秒决策60次的速度玩贪吃蛇！

https://t.co/31KGUiNunb

[Nitesh Seram @niteshseram]

.@grok
What are the top five dumbest things I've said on this account?

[Gayathri Perumal 🥑 @gayathrithedev]

.@grok
What are the top five dumbest things I've said on this account?

[Nitesh Seram @niteshseram]

@vasu2001_ yeah fair. I guess my concern is more about going with the first solution AI gives you because it works, without thinking how else you could’ve done it

[Nitesh Seram @niteshseram]

Is AI a blessing for productivity but a curse for creativity?

When you get working code in seconds, it’s easy to stop exploring other ways to solve the problem.

Are you experimenting more with the time you save, or just moving on to the next task?

[Nitesh Seram @niteshseram]

@vasu2001_ claude is employee of the month now😀

[Nitesh Seram @niteshseram]

well that dream lasted 8 seconds

[Nitesh Seram @niteshseram]

@RachitThakur146 Really cool

[Sam Altman @sama]

There are two ways AI progress could go very badly and that we must avoid.

First, we could lose control of the future to AI. This is unacceptable; we are unapologetically on Team Humanity, and AI must always serve people. To ensure that, we need ways to ensure that alignment and safety techniques stay ahead of progress in model capabilities.

Second, we could end up in a world with too much concentration of power. If an extraordinarily powerful AI is used by one person or company to impress their worldview onto everyone else, the results could be extremely dystopian.

Avoiding these two threats requires walking a narrow middle path; for example, one country could gain too much power. Another example is one lab ending up with too much power.

[Nitesh Seram @niteshseram]

@tiya_decodes says “one small change” and the diff is 2,000 lines

[Nitesh Seram @niteshseram]

my archive button has a puppy in it as well.
now i keep unarchiving things to let him out 🐶

🛠️ @reactjs @motiondotdev @tailwindcss

[Nitesh Seram @niteshseram]

my input won’t stop watching me type.
at least he looks away for passwords 🐶

[Nitesh Seram @niteshseram]

What if bookmarking a page felt like folding its corner?
Built with React, Tailwind & Motion. Sound on 🔊
Would you ship this in your app?

[Nitesh Seram @niteshseram]

now Neki handles routing across all those shards underneath in your app

[Nitesh Seram @niteshseram]

118m qps is wild, but the part that got me is how boring the scaling looks

5 shards → ~1m qps
50 → ~10m
512 → ~118m

for this workload, they just kept adding postgres shards and throughput kept following

kinda crazy

[PlanetScale @PlanetScale]

118,538,803 queries per second.
1.22 PiB of data.
512 shards.
Powered by Neki.

https://t.co/9wAVjQoZ1B

[Nitesh Seram @niteshseram]

now you have multiple Postgres databases, and something has to figure out which one each query should go to.

[Nitesh Seram @niteshseram]

@alexandr_wang first feature request: let me use it in India 😭