Here's mine: maxlavender.com
213
Everyone's using AI to design faster.
I built Hold My Figma with @Bot to tell me when I shouldn't design anything at all.
Give it a product brief and it tries to kill the premise before you waste time solving it.
It interrogates the evidence, hunts for competing explanations and actively tries to prove itself wrong.
Then it makes the call:
🟢 PROCEED
🟠 INVESTIGATE
🔴 STOP
I gave it a brief to improve a broadband package selector.
It came back: don't redesign it yet.
The evidence suggested price comprehension and total cost were stronger hypotheses than the selector itself.
So instead of generating UI, it proposed the cheapest experiment to find out.
The next step became an experiment, not a redesign.
Try Hold My Figma:
x.ai/bot/kcobz9DbN4LfAcAFXIA…
We’re inviting one Grok Bot user to a Starship launch. You could win the invite by sharing the ways you've integrated @Bot into your work. The invite includes a plus one.
How to submit and some examples we love:
113
- Scale economies
- Network economies
- Counter positioning
- Switching costs
- Branding
- Cornered resource
- Process power
Most successful businesses are built around one or two
most of the indie hackers like @levelsio or @yongfook will keep seeing this trend because their apps don’t have any of these moats:
> expensive data
> network effects
> attention
Ahrefs won’t be replaced by AI because the cost of acquiring and managing the data is greater to run solo vs just “renting it” and data is the perfect product for agents = context. In fact it might even become more valuable.
Social media won’t be replaced by AI = network effects + attention.
@marclou has a perfect example that isn’t threatened by AI and I bet he will only see his revenue grow specifically with TrustMRR with data + network effects moat
1
97
Always easy to see a polarising example and think this is the way it should be done. Many and the majority of great companies were founded, and built on in person collaboration. Solving complex problems, in collaboration with other humans at scale. @37signals have done an absolute amazing job, however the work location conversation isn’t binary, and very much context dependent
.@DHH on why he likes being alone and working remotely:
“When I look back upon my career, it only took off when I could close my door and not see anyone for long stretches of uninterrupted time.”
“We have about 60 people. Twice a year we meet up and spend a week together and it's really great. You get to see people and I cherish that because you need that human connection. You need to sit right in front of someone to reestablish trust batteries and realize it's not all avatars and text—it's actually real people.”
“But at the end of that week I'm always like, ‘As much as I love these people, as much as I love this company, I would fucking quit and live as a hermit in the forest rather than work with them 365 days a year in an office.’”
“When I got started working in technology, I worked in open offices. And you know what? My productivity and my creativity was a bit of a flat line.”
64
So much opportunity out there as designer. Yes, craft is important, and naturally, visuals will always get the attention. However, if some of these AI labs/IDE vendors invested more in problem focused product designers, game changer
1
85
“The unique stream of knowledge, signals, and observations you’re exposed to by virtue of where you sit: your geography, network, industry, or vantage point. It’s an edge that comes not from being smarter than others, but from seeing things others structurally can’t — because they’re looking elsewhere, or not looking at all.”
Information Flow¹
—
¹ Charlie Songhurst on Invest Like the Best (2020),
"Lessons from Investing in 483 Companies"
85
This it’s so hard with where “Designer Twitter” was, not that many years ago
“This is why followers and likes are not a great proxy for what’s good and what’s not.
Many people are good at posting shiny things, but as soon as you look for any tangible work, it falls apart. Mainly because building is a completely different skill, and that’s the skill you actually want.”
After my last post about turning someone’s content into a skill, a few people asked the same question: how do you know who’s actually worth learning from, given all the AI slop out there?
Look for proof of work.
If I wanted to learn font design, I’d read everything Rasmus has ever said about fonts, because he made Inter. I wouldn’t search for the most viral typography thread.
Shipped work that strangers pay for or depend on is the best signal there is.
Before taking someone’s advice, open their GitHub or personal site. Find something they built that people actually use: a product, an open-source library, something tangible.
This is why followers and likes are not a great proxy for what’s good and what’s not.
Many people are good at posting shiny things, but as soon as you look for any tangible work, it falls apart. Mainly because building is a completely different skill, and that’s the skill you actually want.
You want to learn from the builders. Be allergic to the talkers.
109
I've seen in the past people create these second brain/mind map UI's, anyone know what is best to do that?
1
1
40
Great example of an agency site; work + contact laid out in a super clean way. Answers the questions instantly "Why should we hire you?" and "How do we get in touch?" 👌
1
78
Credit where credit is due @OpenAI's design team cooked with the ChatGPT/codex app
29
Built a little X photo interaction with @grok 4.5 and @cursor_ai
Tap the grid → open the post → swipe the carousel → pull up comments without crushing the photo. Swipe down and you’re out.
Mostly just wanted an excuse to get the motion feeling right. Grok was surprisingly good at the fiddly bits (sticky header, the comments sheet, getting the dismiss to stop being weird).
Prototype vibes only. Feedback welcome if anything feels off.
@benjitaylor @nikitabier
1
165