Replying to @lightsilver323
"My app, It shits money!"
Aelox.com retweeted
Be careful how you treat others in a situation because you may find yourself in the same spot.
Aelox.com retweeted
Imagine spending your whole life becoming an academic expert. Then a random guy online tells you that you are wrong about your own field. And he's right. But you can never admit that. Because it would mean admitting that your life was a lie. That is the dilemma of many academics.
Aelox.com retweeted
How to win:
- Do what you say you will.
- Learn from your mistakes.
- Never f*cking quit.
So simple, but most won’t do it.
Aelox.com retweeted
The difference between sounding smart and being smart is “I don’t know.
@naval
Aelox.com retweeted
I repeat this often, "All the magic you're looking for is waiting in the work you're currently avoiding."
Aelox.com retweeted
Momentum feels like magic because it changes how you think about yourself. Once you see progress, your fear and doubt loses power. Start immediately.
Aelox.com retweeted
Marketing > sales. You can be a killer closer, but if no one knows you exist, you starve.
Aelox.com retweeted
The people living how you want to live aren’t better than you… they just stopped settling.
Aelox.com retweeted
get off grid
get rid of distractions
gather a team
pick one thing and BUILD
your life and the society you want, depends on it
NOW MORE THAN EVER
Aelox.com retweeted
The faster you move, the more paths appear. Momentum is the currency of new opportunities.
Aelox.com retweeted
The paradox of being ambitious:
Certain about your vision.
Uncertain of how you’ll get there.
Aelox.com retweeted
Brutal truth: Your inability to focus on one thing for an extended amount of time will hurt you far greater than your lack of intellect.
Aelox.com retweeted
Lots of hot takes on whether it's possible that DeepSeek made training 45x more efficient, but @doodlestein wrote a very clear explanation of how they did it. Once someone breaks it down, it's not hard to understand. Rough summary:
* Use 8 bit instead of 32 bit floating point numbers, which gives massive memory savings
* Compress the key-value indices which eat up much of the VRAM; they get 93% compression ratios
* Do multi-token prediction instead of single-token prediction which effectively doubles inference speed
* Mixture of Experts model decomposes a big model into small models that can run on consumer-grade GPUs















