Serverless platforms like AWS Lambda start an isolated virtual machine the moment your code needs to run.
Booting one normally takes seconds. Users wonโt wait seconds.
I wrote a blog on how this is solved. From the basics to the cutting edge of research, one step at a time. ๐
shows why RLHF trained models might be stuck at "guessing what a human would do" level of intelligence for unstructured and ambiguous tasks where there is no clear winner.
shows why RLHF trained models might be stuck at "guessing what a human would do" level of intelligence for unstructured and ambiguous tasks where there is no clear winner.
@yogeshsingla481 it's so easy to make one, i can't stop making them. so here's another one which took 2 prompts and <10 mins. Try for yourself!
Timber Reflex
I made this game in under 5 minutes using Gemini Argon.
and yes it runs on your phone
Today I'm launching Wish Play โจ
https://t.co/JOhHRl394m
๐ฎ Scroll the feed just like reels and play games made by people around the world, one swipe at a time
โ๏ธ Create your own game in minutes: just describe it in plain words
๐ Show it to the world: publish it to the Feed and let anyone play.
No code. No engine. Just your idea.
Free, live now
Serverless platforms like AWS Lambda start an isolated virtual machine the moment your code needs to run.
Booting one normally takes seconds. Users wonโt wait seconds.
I wrote a blog on how this is solved. From the basics to the cutting edge of research, one step at a time. ๐
I made this game in under 5 minutes using Gemini Argon.
and yes it runs on your phone
Today I'm launching Wish Play โจ
https://t.co/JOhHRl394m
๐ฎ Scroll the feed just like reels and play games made by people around the world, one swipe at a time
โ๏ธ Create your own game in minutes: just describe it in plain words
๐ Show it to the world: publish it to the Feed and let anyone play.
No code. No engine. Just your idea.
Free, live now
@zekramu the average is going up simply cause they're not hiring for junior devs anymore. it's all senior and staff vacancies with high comp on any careers portal
Just learnt about Weighted Rendezvous Hashing, a stable hashing algorithm which is particularly useful when you have servers whose capacity changes frequently. This is the specific case where consistent hashing typically falls short in terms of data locality.
Now even when weights or capacities of servers change, WRH isn't affected in terms of data locality in contrast to consistent hashing. It is a unique intersection of perfectly stable hashing with precisely weighted stable hashing.