Brady Price

ai obsessed - founder at

[Brady Price @BradyPrice91]

@anthonyboutall I do not like Trump and somehow that answer makes me like him more. I don’t relate. But something about the level of authenticity is admirable.

[Brady Price @BradyPrice91]

@ClintT_land Uh oh. This is the opposite of what I thought. How does that work?

[Brady Price @BradyPrice91]

@dvyio @thsottiaux Happy dev day 😂

[Bright Mirror @_brightmirror]

Made with Claude Opus 5.5.

The fearmongering about AI always makes us forget that NOTHING WENT FOOM, as was always predicted.

Send this to your doomer friend who has a very high P(Doom).

Accelerate.

[Brady Price @BradyPrice91]

@BeingAshwin https://t.co/O7knxDZTB9

[Brady Price @BradyPrice91]

@jasonugc SYSTEM

[Brady Price @BradyPrice91]

@mattpocockuk I’m reluctant to add anything like this from other creators, or my own, but I would trust yours.

[Brady Price @BradyPrice91]

@ivanburazin Use raycast

[Brady Price @BradyPrice91]

@thsottiaux Don’t require 2fa to use mobile connected to codex

[Brady Price @BradyPrice91]

@mattpocockuk I used warp for 6 months but ive started using conductor more and more.

[Brady Price @BradyPrice91]

@BraedendotTECH You really need to use a tool that doesn’t require you to use separate terminal windows and you really need to use auto mode or dangerously skip permissions. That way your brain can melt much more efficiently.

[Brady Price @BradyPrice91]

@gsivulka Truly can't relate.

[Brady Price @BradyPrice91]

@mattpocockuk I ignore the code for a while and then point a sota model at the codebase and have it read your skills or books like asopd for several iterations and make it better. Gpt looped for 3.5 days on my codebase finding seams, breaking them, and writing tests.

[Brady Price @BradyPrice91]

@mattpocockuk Imo, invisible losses are too costly and 2% better or worse judgement over many decisions compounds. If the value you get from the tokens exceeds their cost, then I don’t care. I’d rather throw more compute at a problem and save time.

[Brady Price @BradyPrice91]

@myprasanna @vorflux @Rippling Looks pretty incredible. My bottleneck is iterating quickly when building apps that are Agent native and updating my eval harness to keep up.

[Brady Price @BradyPrice91]

@elvissun I dont think the demo gif helps. I actually had to spend a couple minutes understanding the difference and which you were saying is better. BUT a dedicated "our ai vs their ai" carries a lot of weight. I would read that description if evaluating products.

[Brady Price @BradyPrice91]

@elvissun Found you yesterday. I spent all day on your article and I've come back to this tweet a few times. Can't wait for another article with more detail. 🙏

[Brady Price @BradyPrice91]

@omarsar0 I go even further and limit my grok spend to $0. Huge unlock.

[Brady Price @BradyPrice91]

@mattpocockuk Cook!

[Brady Price @BradyPrice91]

@Steve8708 Why would you use this instead of having the agent do it via Agent browser or playwright? If it’s for the commentary, I can dictate my commentary to the agent, and it can be far more thorough in reproducing the bug and checking edge cases.