ethan /entropy

You know what you want to get done. Now your AI can too, with minimal time, tokens, and tending.

@DIUXdsgn @galshirart @Labrys_io TIL taste = ♻️

@galshirart @Labrys_io Them ai grifters will never reach this level of taste

Logo design for @Labrys_io ✨

@Zeneca (not willingly btw)

@Zeneca recently heard that ~30% unemployment is needed to offset the oil supply disruptions from the war, with potential recovery looking like ~2029

I agree, have been feeling this way for a while I think it gets worse before it gets better

@pwnies Well this is real fun. @emileevee diffui really wants to make a catmorphic version of u

I'm launching something new on diffui - the ability to define brands. Brands will let you upload reference screenshots, from which we'll build out color/typography/design system assets and usage guidelines. With brands you can create any design in your own brand language.

Hey so with all the AI / crypto news recently, just wanted to remind y’all: >> not your keys, not your tokens << or are the 12 words too hard to manage yourself now that you let llms think for you all day??

“Non-technical teams are now shipping production code”

@rileybrown A very smart SWE just echoed the same sentiment to me when inquiring… 🤔

I hardly ever say this... but my spidey senses are tingling... something is off about this. mark my words.

@anvisha Oooh this is going to be so great in my workflows!

Introducing the Moda API: the design layer for agents, apps, and workflows. Embed presentations, reports, and design creation right into your product or agent. See what people are building with it 👇

@Everlier @Polymarket JUST IN: custodial wallet holders after this

@Polymarket JUST IN: the first capture of such a team live at work

JUST IN: Coinbase to test AI-native “one-person teams” that combine engineering, design, & product roles.

I’ll start- I’m building a true day off into my schedule (soon) ✌️🌲

GM, what are you building today? (wrong answers only btw)

@wabdoteth seals are one of my very few remaining loves of the last few years stay sappy 🫰 arf

Over the years, we’ve made a lot of changes to the brand, the logo, the characters, the art style, and everything in between, but never really formalised any of it until now. That was because I was never fully 100% happy with what Sappy was at any given moment; there was a massive divergence between what Sappy was, and what *I* knew Sappy could be, and commiting to those earlier versions felt extremely premature. Ossifying these things didn't feel like the correct thing to do, especially when I knew deep down that they would have to be aggressively scrubbed away or retconned (which gets really hard to do once a brand starts to genuinely penetrate and build momentum). There was also a divergence between what Sappy started as, what other people wanted it to become, and what I wanted it to be. Early on, I did see Sappy as more of an art project than a traditional "brand", which is why I leaned so heavily into memetics at a time when practically nobody else was doing it, and why I treated the character more like something alive and culturally reactive--much closer to how people think about memecoins nowadays and not too dissimilar to performance art. I also hated the word “brand” and the corporate aura it brought to the community & surrounding subculture. Being honest, I still think there’s something correct in that instinct. NFTs derive most of their speculative value (aside from the occasional airdrop narrative) from the community and the interconnected social clusters around them, not from “IP growth” in the classical sense. At the same time, it felt like there was enormous pressure (especially in comparison with other PFP projects) from speculators to turn the Thing into something bigger and larger scale than it is right now, because in startup [& crypto] culture, growth is treated as the ultimate metric, and it was applied directly to NFTs despite mostly being antithetical to the IP business. Internally I wanted to avoid this, slow down and take time to figure out what the brand should be, which sounds like total crazy talk in the crypto world because there's a lot of pressure (rightfully so) to grow grow grow, at all costs. We still did it of course, because I'm way too competitive of a person not to, but I think it was very evident from the outside that my heart really wasn't in it (at that point in time, chasing those specific reasons). My dopamine receptors might be fried, but I still find directing/creating a meme that people genuinely find funny more rewarding than watching a random reel of ours get 40M views or having billions of GIPHY views or whatever other metric sounds good to say out loud. What makes this all so strange is that NFTs are hyperspeculative assets that need constant cultural fuel to keep the surrounding network expanding faster than the floor price alone would justify. That model is in many ways at odds with how truly loved brands are formed. Real brands are much more malleable, they take time, they evolve; they’re not meant to be frozen in place on day one, and they're not meant to be subject to a quorum when more drastic changes that disturb the status quo are made. Some of the most adored brands globally took years, or even decades for the artist/creatives to actually hone in on their essence that was then expanded upon and delivered to the world. And I think to some degree, crypto and NFTs create such insane mental dissonance for builders. They train you to focus on things that aren’t actually important, usually through confirmation bias or false signals--of course it would; when you look around and see some random builder making some logically illogical absurd Thing that somehow (unsustainably) prints tens of millions overnight, it becomes very easy to convince yourself that the weird Thing is the thing that works. That's the thing that finally cracked the crypto formula, that's the model that works, that's what with a little iteration could lead to the golden goose that every retail investor, VC investor, and every crypto founder has been chasing since the first smart contract was created. It's somewhat embarassing to admit, but I spent years of my life believing that "tokenomics" was a real thing or that /if you just tweaked [insert random weird crypto-native variable] just slightly enough it could work/. It's different now that most of the speculative juice has left the industry and everything has deflated, clarity is much more abundant now. Another embarassing thing to admit is that a lot of the things we are doing for Sappy now are things I would have done years ago...if the brand was founded outside of crypto, and if I listened to my gut earlier. There really is no excuse for that, other than psyopping myself into believing that there was some special "model that can work for NFTs", or that building a crypto-founded character had to take a non-traditional route, or that what peers were (also) currently doing was the correct way to build, and not just appeasing a shrinking echochamber whilst playing the ponzi token fugazzi game. I wish I could have learned these things a little faster, but I also did start out as a tech bro undergrad that knew next to nothing about brands or art or IP outside of random video games I grinded. I didn't participate in consumer culture or fashion or anything similar at all, I knew next to nothing about business outside of briefly being into crypto/blockchains, and was purely operating off of my gut feelings and what seemed correct to me. Arguably that gut instinct was pretty precise, but the inexperience & uncomfortability is what led to not being able to pull the trigger on certain things or execute in the sharpest way possible. In truth, there is no other way it could have gone because most of those things had to be experienced and there is no real way for it to be crash coursed. This post was supposed to be a two-liner about how I'm redoing the brand properly from the ground up, with unique aesthetics, a cohesive universe with real character development, comics, longer form content (thanks to AI), sharpening every touchpoint, creating proper funnels to guide normies to our products and become buyers/users, but it ended up becoming this long-winded explainer of how we even got here in the first place. Which is a little poetic, because the path to the end is never straightforward. [Also sorry to rug you but this logo isn't actually a final logo I just like many aspects of it and what it represents]

@Antification ngl i’d throw money on this one just because i support the use of free will to annoy and take the narrative power control away from them

What’s the ca guys

@poetengineer__ cc @lorepunk

a perhaps better model for ai writing: instead of me prompting the llm to write, the llm should prompt me to write. just like in poetry workshops the instructor simply gives meaningfully crafted constraints: write a sonnet, now write a haiku, and now without adjectives.

@threadreaderapp @StartupYou @jackclarkSF eww. clicked this expecting to find a new useful tool, but there’s so many fuckin ads i immediately closed it 🙅

@StartupYou @jackclarkSF @StartupYou Hallo, you can read it here: threadreaderapp.com/thread/2051312… Talk to you soon. 🤖

I've spent the past few weeks reading 100s of public data sources about AI development. I now believe that recursive self-improvement has a 60% chance of happening by the end of 2028. In other words, AI systems might soon be capable of building themselves.

@M_S_Rabi bro put that away respect the hype centerpiece

Perfect execution and phony vision - meaning the founder has taste but not the depth. Good luck though

@skeptrune a lot of startups are just good at twitter. full stop

a lot of startups is just being good at twitter everyone assumes things are going terribly the moment you stop being relevant on this app