Sebs

130 posts

Sebs banner
Sebs

Sebs

@Sebastian98MM

Designer, Engineer, Builder

Katılım Mayıs 2012
128 Takip Edilen21 Takipçiler
Ry
Ry@heyhaigh·
I was lucky enough to work with the @supdotxyz team for the last 5 months. They're leading the way in rebuilding an internet that we can all call our own again; the web that was promised. Bash is that path forward. I have some invites for direct connects who want to give it a go. Give a shout if interested.
dom hofmann@dhof

excited to share a new toy called bash. it’s a multiplayer coding agent with social built around it starting with invites and a waitlist, and we’ll open it up as quickly as possible bash.tv

English
2
0
8
869
Sebs
Sebs@Sebastian98MM·
@figma elite ballers use the track pad and mouse combo
English
0
0
3
475
Figma
Figma@figma·
If you use a trackpad who hurt you
English
210
77
900
210.3K
Sebs
Sebs@Sebastian98MM·
@thinkwithmark ofc this is the type stuff I was hoping for with ai!!
English
0
0
1
16
Sebs
Sebs@Sebastian98MM·
I’ve tried whisper flow and enjoy but man idk it just never stuck with me even though it would probably help with thinking through my ideas. Although I tend to be able to articulate better through text so that’s always been a bottleneck. But this this I think really does something different I read a lot of articles and being able to save my notes instead of having to go back and forth between screens would be great.
Mark™@thinkwithmark

We raised $1M dollars to reinvent how people read. Introducing Mark II - a $159 AI bookmark. Thread below

English
1
0
1
144
Agatha Richards
Agatha Richards@thedesignely·
Help, which mask reveal looks better?
English
7
0
14
613
Sebs
Sebs@Sebastian98MM·
Cotejo v2 coming soon 🐈‍⬛
Sebs@Sebastian98MM

I reopened my wip portfolio a while back because I wanted to redo the type. Simple, normal designer spiral. But then something funny happened — I realized I actually loved the space the old type was taking up. The rhythm was right. The air was right. The letters were holding the page in this really nice way. I just… did not like the actual typeface anymore lol. So I went looking for a replacement, and that’s where I got annoyed. I didn’t want a font that was “better” in some vague taste-board way. I wanted a font that could drop into the same exact footprint as the old one. Same space. Same presence. Different voice. And basically every tool made that weirdly hard. You eyeball it. You resize it. You guess. You pretend it’s close enough. So I built Cotejo. Cotejo means a careful side-by-side comparison in Spanish, which felt right because that’s the whole point. It cap-matches every candidate to the typeface you already have, so everything shares the same cap height and fills the same space. You’re not judging size anymore. You’re judging the actual letterforms. A type swap should feel like it slots in, not like you’re wrestling the page back into place. It also recommends font partners across your own library and Google Fonts, lets you fine-tune pairings on a tracing-paper-style overlay, and exports real CSS / tokens when you’re done. @joshpuckett — Pica and Locale were a huge north star for this. That quiet, instrument-grade kind of craft where the tool gets out of the way and the type becomes the whole thing. I kept thinking: this is such a small, specific problem, but if you’ve been stuck in it, it is painfully real. Would genuinely love your eyes on it. cotejotype.vercel.app github.com/SXM4434/cotejo

Español
0
0
0
22
Sebs retweetledi
Sebs
Sebs@Sebastian98MM·
I reopened my wip portfolio a while back because I wanted to redo the type. Simple, normal designer spiral. But then something funny happened — I realized I actually loved the space the old type was taking up. The rhythm was right. The air was right. The letters were holding the page in this really nice way. I just… did not like the actual typeface anymore lol. So I went looking for a replacement, and that’s where I got annoyed. I didn’t want a font that was “better” in some vague taste-board way. I wanted a font that could drop into the same exact footprint as the old one. Same space. Same presence. Different voice. And basically every tool made that weirdly hard. You eyeball it. You resize it. You guess. You pretend it’s close enough. So I built Cotejo. Cotejo means a careful side-by-side comparison in Spanish, which felt right because that’s the whole point. It cap-matches every candidate to the typeface you already have, so everything shares the same cap height and fills the same space. You’re not judging size anymore. You’re judging the actual letterforms. A type swap should feel like it slots in, not like you’re wrestling the page back into place. It also recommends font partners across your own library and Google Fonts, lets you fine-tune pairings on a tracing-paper-style overlay, and exports real CSS / tokens when you’re done. @joshpuckett — Pica and Locale were a huge north star for this. That quiet, instrument-grade kind of craft where the tool gets out of the way and the type becomes the whole thing. I kept thinking: this is such a small, specific problem, but if you’ve been stuck in it, it is painfully real. Would genuinely love your eyes on it. cotejotype.vercel.app github.com/SXM4434/cotejo
English
0
1
1
261
Sebs
Sebs@Sebastian98MM·
@jenny_wen @cursor_ai Generational run Figma anthropic and now cursor. Your talks what finally got me to try Claude maybe 🤔 I gotta give cursor a try
English
0
0
0
1.4K
jenny wen
jenny wen@jenny_wen·
ok! quick updates from me: - i left anthropic (who does that?!) - had a baby (i love her) - and am joining @cursor_ai as head of design (eep!) it's been a low-key dream of mine to nurture a team that cares so deeply about craft, quality, and building great tools. very excited!
English
712
202
10.4K
1.3M
Eli Heuer
Eli Heuer@eliheuer·
@DannPetty I am, I have a new blog post coming out monday about how I use img2bez, a custom harness, and a small neural network I trained to do type design. I will try to get an interview next week after it is published.
English
1
0
5
307
DANN©
DANN©@DannPetty·
Your portfolio doesn't need more projects. It needs one project explained like you actually thought about it. I've reviewed thousands and I can count the ones that did this on one hand. Quality over quantity.
English
18
8
375
19.5K
Sebs
Sebs@Sebastian98MM·
@poetengineer__ Everyone keeps telling me I’d love learning touch designer it’s one creative tech tool space I haven’t touched
English
0
0
0
83
Kat ⊷ the Poet Engineer
Kat ⊷ the Poet Engineer@poetengineer__·
𓆝 𓆟 𓆞 𓆝 𓆟 experiments with instancing along a 3d curve
English
28
131
2.1K
61.1K
Sebs
Sebs@Sebastian98MM·
@jshguo cotejotype.vercel.app it’s a type comparison tool that helps you compare fonts. It great for when you like teh space a font takes up but don’t love the feel and look.
English
0
0
1
50
Joshua Guo
Joshua Guo@jshguo·
Alright, the first version is almost ready. I’m curating a small archive of tools, experiments, prototypes, and interfaces made by designers using agents. Have you built anything interesting with agents? Or seen something worth adding? Drop it in the comments.
Joshua Guo@jshguo

Thinking about building a curated archive for designers building with vibe coding. Tools, visual experiments, data visualizations, prototypes, tiny products, and interfaces that feel human-designed, amplified by AI. Would this be useful?

English
1
1
12
2.8K
Sebs
Sebs@Sebastian98MM·
@blaketshao Im surprised it took so long you could this in 3d software through scripts and much of my first early learning with code as a kid was with node based stuff
English
1
0
0
16
Blake Shao
Blake Shao@blaketshao·
@Sebastian98MM i think it's the ethos of build bespoke tools for specific projects, and node system makes a great medium!
English
1
0
1
67
Blake Shao
Blake Shao@blaketshao·
call me a psycho, but sometimes I really really really want to make graphic designs programatically this is my take on a node-based graphic design tool. something about seeing the design grow itself once I built the pipeline #nodepilled
English
16
20
415
21.2K
Sebs
Sebs@Sebastian98MM·
@figma hey I’ve contacted support almost 5 days ago my account sharing has been restricted all the ai agent errors out and trying to publish or edit any off Figma makes have been disabled. Would be grateful if I can get some help much of my work is done in Figma and as a freelancer this makes it difficult for me.
English
10
0
1
131
Sebs
Sebs@Sebastian98MM·
haha yes, even with software engineering it can be the samecthing. I was a computer engineering major but first half of college then went to design . From what i remember from the cs and engineering classes i took there is def just a big difference between the two. I would say my cs classes where much more similar to many of the design courses.
English
0
0
1
43
joshpuckett
joshpuckett@joshpuckett·
@Sebastian98MM @lochieaxon I generally think that real engineering fields (e.g. with licensure, standards, etc) have far more defined scopes than tech. E.g. where does a 'design engineer' end and 'front-end' or 'product' engineer begin? No one knows..
English
1
0
1
178
joshpuckett
joshpuckett@joshpuckett·
TGIF! Closing thoughts on "Design Engineers" for the week: Part of the title's popularity, I think, is that it's a new way to be the last person in line to take credit. @lochieaxon and I were talking about this trend earlier, how it's like when the waiter gets the praise for the wine being good. (1) Over the past couple years, thanks to AI, we're seeing a new crop of maker emerge that presents work as if they did it all. Designed it and engineered it. This of course does happen, and many companies value people who can do this. But more often than not, great work is made through collaboration. Wonderful folks like @jakubkrehel and @emilkowalski collaborate with @paulfaivret and @glennui, respectively. At the end of the day, all that matters is doing the best work possible and collaborating as much as you need to achieve that. Focus on that, and don't worry about what you're called. (2) ----- 1. Lochie is criminally under-appreciated and refuses to take credit for his very meaningful contributions to our field. He's a class example of how to hold a relentlessly high bar while deferring all praise to your team. 2. Since deciding to exist publicly on the internet again this year, many folks have labeled me a 'design engineer'. I always chuckle at this, as I just think of myself as a designer despite coding my whole life.
joshpuckett@joshpuckett

Ok happy Monday! At risk of giving a slightly trope-y response, I think of design as the ability to identify and solve leveraged problems. I think of engineering almost exactly the same as design, just with more emphasis on the solving part. I see many people who are adopting the persona of “design engineer” who seem mostly into the veneer layer of software, like making interfaces a little more animated or expressive. Those aren’t in and of themselves bad things necessarily, but if all you are doing is taking someone else’s design (their solution, their intent, maybe exemplified in a facsimile like a mockup) and implementing it into an existing system (following established principles and frameworks, never really concerning yourself with anything beyond the DOM), then that’s just like being an assembly line decorator. Which doesn’t have zero value! But it has far less value than someone I could task with figuring out how we should increase a specific KPI or take better advantage of caching and bundling to reduce costs and load times for our mobile users if we are blowing up in parts of Asia or Africa and understands the tradeoffs of multiple competing solutions. So instead of chasing trendy titles I think it’s more valuable to just excel at design or engineering, first, before trying to do both and, like most fusion restaurants, just being bad at either.

English
11
8
125
14.5K
Sebs
Sebs@Sebastian98MM·
@gabbisoong oh i am not 😅. type is something i love but i struggle with. i just well was struggling with pikcing fonts for my site so i ended building a more scrappy version of this and thought it could help people!!
English
1
0
1
20
gabbi
gabbi@gabbisoong·
@Sebastian98MM haha woahh this is so cute! i love the onboarding, are you a type designer?
English
1
0
0
59
gabbi
gabbi@gabbisoong·
"Every major scientific breakthrough relied on a visual tool that made something invisible visible. Darwin needed side-by-side illustrations of finches to see variation that was otherwise too subtle to notice." even if chatgpt can tell you "which fonts are similar to inter", we wouldn't truly understand until we can literally compare these fonts visually side-by-side (hence why arcotype is a map, not a list)
gabbi tweet mediagabbi tweet media
Yasmine Khosrowshahi@yasminekho

Stanford professor Judy Fan went on stage at MIT and broke down why humans are so good at making the invisible visible... And why AI hasn't actually learned to "see" the way we do. It completely changes how you think about Human Intelligence v/s Artificial Intelligence: 1. Nature never gave us straight lines or sharp corners. The number line, the coordinate plane, even basic geometry are all human inventions. We created tools that do not exist in nature simply because we needed a way to think more clearly. 2. The coordinate system Descartes invented solved a problem that had stumped mathematicians for centuries, doubling the volume of a cube. Once invented, this tool became so indispensable that virtually every math curriculum on Earth still depends on it. 3. Humans have been doing this for at least 30,000 to 80,000 years. The story of human progress is inseparable from the story of marking up our environment, from cave walls to Galileo's telescope to Feynman diagrams of particles we will never see with our own eyes. 4. Every major scientific breakthrough relied on a visual tool that made something invisible visible. Darwin needed side-by-side illustrations of finches to see variation that was otherwise too subtle to notice. Cajal needed detailed drawings of neurons under a microscope to map how the nervous system was wired. 5. Fan's research group studies something deceptively simple: how people decide what to put into a drawing and what to leave out. When two people played a drawing game, sketchers used far more detail when the target object had close competitors than when it stood alone, all the way down to using fewer strokes and less time when more detail was not necessary. 6. People are not just copying what they see. They are making constant judgment calls about what level of detail actually serves the goal of communication, and they do this naturally without ever being taught the theory behind it. 7. There is a real difference between drawing something so someone can identify it and drawing something so someone can understand how it works. In one study, participants drew explanatory diagrams that emphasized moving, causal parts of a machine while depictive drawings emphasized background and overall appearance, even though both were drawing the exact same object. 8. Explanatory drawings were genuinely better at helping someone figure out how to operate a machine, but worse at helping someone identify which machine it actually was. You cannot optimize a single drawing for both goals at once. Communication always involves tradeoffs. 9. AI vision models trained on photographs generalize surprisingly well to simple, sparse sketches, suggesting that resemblance based recognition is not just a story we tell ourselves. It is something modern neural networks can replicate with real accuracy. 10. But there remains a large, measurable gap between how confidently AI models recognize sketches and how confidently humans do, even when both groups answer the same questions about the same images. Humans are simply far more reliable and far more consistent in their judgments. 11. When researchers compared human-made sketches to AI-generated sketches under tight stroke budgets, both were similarly recognizable at higher budgets, but diverged sharply as the budget shrank. Humans and AI systems simplify drawings in fundamentally different ways once resources get scarce. 12. Reading a graph is not one single skill. It involves perception, knowing where to look, mapping that visual information onto the actual question being asked, and then translating that mapping into an answer. Each of these steps can independently break down, and people fail for very different underlying reasons even when they land on the same wrong answer. 13. When tested directly against humans on graph reading tasks, leading multimodal AI models, including GPT-4V, showed a meaningful performance gap. Even when a model's overall accuracy approached human levels, its pattern of mistakes looked nothing like how humans actually get things wrong. 14. People choose entirely different types of charts depending on what specific question they are trying to answer, not out of a generic preference for bar charts or scatter plots. Their chart choices closely tracked which visualization would genuinely help someone answer that specific question correctly. 15. Two of the most widely used graph literacy tests in education research turned out to correlate strongly with each other, suggesting they measure overlapping skills. But when researchers dug into the actual error patterns, the standard categories used in textbooks, like "find the maximum" or "identify a cluster," failed to explain why people got things wrong nearly as well as a more basic, underlying four-factor model did. 16. The deepest goal behind all of this research is not just academic curiosity. It is to eventually help students and everyday people develop genuine literacy with the visual tools that science and modern decision-making increasingly depend on, because every generation should be able to see further than the last by standing on the visual tools the previous generation built. Follow @yasminekho for more ideas on thinking better, becoming clearer & building a more intentional life.

English
4
3
107
10.5K
Sebs
Sebs@Sebastian98MM·
I like testing new AI tools the dumb/simple way first. No perfect prompt structure. No fancy domain vocab. Just type the first thought and see what it gives me. Did that with Figma Motion for a simple animation and honestly… it got me like 85% there. The main struggle was secondary animations and sequencing. It understood the general vibe, but once things needed to happen after each other, or react to each other, it started getting messy. That’s where I lost the most time. I purposely didn’t storyboard upfront because I wanted to test the low-effort ceiling first. Like how far can this thing get when I don’t baby it? Pretty far, actually. But the rough part was making broad timing changes. I’d say stuff like “move everything after this sooner too” or “update the whole sequence around that,” and it kind of understood what I meant, but didn’t fully restructure the timeline cleanly. Also noticed the way your frames/layers are set up matters a lot. Sometimes it would edit the frame one level up, but not go deep enough into the actual nested pieces I wanted animated. So your file structure can either help the AI or completely confuse it. That’s the part where I really wished I could just reorder keyframes, drag timing around, or manually fix the sequence myself. Still cool. Still useful. But yeah — these tools get you close fast, then the last 15% is where the time goes. #figma #figmamotion
English
0
0
0
152