Anish Paleja

@jsonphile Haha, appreciate the support Jason!

@Anipaleja This is why you should use Brevitas

Everyone's flexing token usage. I think in 12 months the flex will be how few tokens your stack needs to get the same result.

@KayaHickin 100%. Right now it's experimentation > efficiency. Once AI moves deeper into production, token efficiency becomes infrastructure instead of a nice to have option.

Valid take, especially for larger enterprises. But today, token usage is still a strong proxy for how much you’re actually building and experimenting. That’s why it’s such a high-signal flex right now. In 12 months, though, I think you’re right. Token optimization will be non-negotiable for people flexing or not.

@soleil_colza_ Thanks Hina!

@Anipaleja Good statement

@pupscub Haha Thanks Adi 🔥🔥

@Anipaleja I think you’re right

It sounds obvious once you notice it. Two machines are talking through a language designed for humans. I don't think that's where AI infrastructure ends up. That's why we built Brevitas: brevitassystems.com

One thing I've noticed is that people treat every token as equally valuable. They're not. A token carrying new information is useful. A token repeating something the model already knows is just overhead. That's exactly the problem we're building Brevitas to solve. If you're building AI agents, check us out: brevitassystems.com

People keep asking which model is cheapest. I think the better question is: why are we sending the same context over and over in the first place?

We spent decades optimizing how computers communicate with each other. AI is one of the first major computing platforms that's still passing around giant blobs of human language. I don't think that'll last.

@daleverett @polygres For sure @daleverett 👀👀

@Anipaleja Wait til u use @polygres and get an effectively infinite context window

@dayumsam22 Thanks!

@Anipaleja You are right!

@LoupWang Appreciate it!! 🚀🚀🚀

@Anipaleja Let’s go!