not enough people are talking about the ChatGPT projects feature.

it's the most underrated AI feature and the easiest way to create specialized experts.

i used it to get in the best shape of my life.

here's how you can use it to help you achieve any goal 🧵

(bookmark this) what are ChatGPT projects?

projects let you preload instructions and link files.

any time GPT responds to a chat within a project, it reads the instructions and files before every reply.

These MCPs are cheat codes that make Cursor "just work."

Here's how they eliminate errors, save time, and improve your app:

(Bookmark this for later)

1. Magic MCP

Not only does it upgrade your UI and animations, it shows you a preview of variations before coding it. Just type "/ui" in the agent chat, describe what you want (re)designed, and instantly get polished UI variations to choose from.

Simple as that. It's unreal.

And because it shows you a preview, you'll be reassured that it will get it right without a tedious back and forth.

After thousands of hours in Cursor, I've found the perfect workflow for vibe coding with no errors.

And no, you don't need to know anything about code.

Here's everything you should do (bookmark this):

1. DO NOT tell it what to do. Provide goals, not specific commands.

Unless you can code, odds are you won't know the right instruction to give the agent.

Give it a problem statement and outcomes. Then ask questions:

> How would you make this?
> What do you need from me?
> What are your blind spots?

i built an app with ZERO coding experience using AI.

it was the happiest day of my life.

here's why i don't recommend anyone doing the same.

my learnings and reflections:

🧵 for the last 657 days, i've designed at least one app or feature idea and posted it to twitter.

657 straight days, during holidays, weekends and through illness. i've had lots of ideas.

the designs have always been meant for someone else to build. because, well... i can't code.

my top 10 apps on ethereum this year 🧵 2023 gave us more creative ways to use the chain. we saw a glimpse of what a thriving consumer crypto ecosystem could be.

that said, existing wallets are the status quo imo, and i've decided to exclude them to showcase what's on the horizon for meaningful crypto use cases.

i'm blown away by how creative websites from the 90s were, despite the constraints.

with slower internet speeds, websites needed to be more efficient, prioritizing function over style. but yet, they still managed to be useful and have flair.

i've curated 15 of my favorites👇🧵 1/ google, 1999

the OG minimalist. once you the others below, you'll better appreciate how ahead of it's time this ux was.

AI won't just replace interface designers. It will replace interfaces.

A vision for digital products in the age of AI 🧵👇

1/31 2/ I'm often asked how AI will change design. There's a lot to contemplate:

What design skills are future-proof?
How will AI augment a designer's job?
What parts of design will be automated?
Will AI replace designers altogether?

Truthfully, these are the wrong questions to ask.

Take a break from doom scrolling and enjoy some brain candy.

Here are 16 oddly satisfying "pull to refresh" animations.

🧵👇 1/ by @guidorosso

AI won't just replace interface designers. AI will replace interfaces. We're moving from buttons to chat. But final form is no input at all.

Think about Nest. Perfect UI/UX: It has a smooth rotation, satisfying haptic feedback, clear temp indicator, beautiful colors.

But you never dial it. It already knows you like it colder while you sleep.

Concept: Crypto x AI x Commerce

Ask a shopping agent for a recommendation and give it permission and guidance to purchase things on your behalf. In this approval scenario, which I think could be really interesting for multi-account/wallet users:

1. Ask for a recommendation, it makes a suggestion based on your specific interests that it has observed about you.

2. Ask it to make a purchase on your behalf by monitoring a… twitter.com/i/web/status/1…

Onboarding the next billion users with words.

How the term "wallet" is psychologically limiting web3's growth potential, and a case for abandoning it.

🧵👇 1/ When you hear the term "wallet," you probably (pre-crypto) imagine a foldable pouch to store paper money.

It's a useful metaphor—crypto has solved many finance problems: store of value, borderless transactions, trustless financial services.

But crypto has become so much more

21 design fails that are so bad, they're good 🧵 1/ Which way is up?

Concept: Online courses gated by enrollment NFTs.

🧵👇

1/16 2/ This is a follow on from yesterday's "luxury software" thread, as a deeper exploration in positive ways to leverage limited access software.

https://twitter.com/0xDesigner/status/1651240773554151429?s=20

Concept: Luxury software

In the near future, the most popular and sought after apps will be the ones you can't have.

A short thread 🧵

1/X 1/ For the last two decades, the web2 playbook for software success has been SCALE:

Find product-market fit, acquire as many users and eyeballs as possible, figure out how to monetize later.

Probably the best thing you'll see today.

In 2017, a group of developers hilariously competed for who could create worst volume control interface in the world.

The results 🧵

1/22 1/