Canโt believe Iโm closing in on the final stage of my pop-a-shot build. Seeing it come to life is a surreal feeling.
The ups and downs were all worth it. Long nights working, stressing, executing.
Finally will be shown to a $20M company.
The amount I learned from just doing it and seeing what happens along the build journey taught me so much.
God speed.
I 100% agree. The quality is not noticeably different anyways, so saving that 30% on tokens is better. Now if you can maximise both youโll do much better work where you bring in opus on the heavier task and then continue back in sonnet is really how you can elevate both quality and token saving.
@Ashutosh_Sidh For entrepreneurship this is gold. Nobody has excuses anymore, it has never been cheaper to have startup like it is now all thanks to AI and the ability to program anything.
Claude Sonnet 5.5 is going to make building better. Itโs a clear upgrade over Sonnet 5, runs more than 30% faster, and costs up to 30% less for most work.
Your telling me I can build more efficiently for a cheaper cost?
At first it was all about building insane models. Now its how can they make there current models cost less for there daily users with the same amount of quality.
@Zaydenf9 Users matter more imo. The reason for that is if you get lots of users that means you will get the revenue. Only difference is users can bring more users on board. Word of mouth is the ultimate marketing strategy.
If you silence the noise and you really really focus on understanding how to use LLMs you can create literally anything you can think of.
Whatโs helps me along my journey with zero computer science knowledge is how to understand what i am asking, learn what the outputs in receiving, and testing the code to make sure itโs actually quality.
Put your head down and grind. You will learn as you walk the path more than you think.
Cloud sessions are officially available and out of research preview! They let you keep Claude Code working, even when your laptop is closed.
Existing subscribers get a one-time credit to try them: $100 on Pro, $250 on Max.
Claude Opus 5.5 has been on fire 5hrs of coding today and burned a lot less tokens specially on the small tasks.
Like I said before they are going to enhance the models they have now to make them more and more user friendly.
Soon all the models we are seeing now will be just as good as each other and the only core difference will be the cost per M token.