Profile avatar
amalgam.bsky.social
@[email protected] 🇸🇪 Swede in Hanoi 🇻🇳 Design, UX, Product and a little bit of AI —with a side of life’s quirks. Posts mostly in English, sometimes in Swedish. https://rosenqvist.design/
240 posts 278 followers 1,623 following
Prolific Poster
Conversation Starter
comment in response to post
I mean… it seems like machines are making plenty of mistakes
comment in response to post
Since I live in the ”wrong” time zone I feel like I’m missing a lot of the interesting posts in a chronological feed. I would love to have an algorithmic feed that could surface interesting topics for me. Bonus if I can configure it myself
comment in response to post
Thanks for writing a review!
comment in response to post
I could see it working if you focus on creating a very good ui/UX for a particular niche of customers. Would need to be a niche that can support your costs though… you are unlikely to scale to a huge corp with that strategy but maybe you can create a nice small company
comment in response to post
Yeah, flawless is something you say to the press while working towards another quality metric in house. He’s skill must lie in the political game he must play to delay a release until it is ”good enough”. It can’t be easy to delay when everyone is expecting it to release before the hype ends
comment in response to post
Also it helps with engagement metrics. The companies need to demonstrate growth to investors who are funding this. Starting each response with ”great question!” helps improve those metrics
comment in response to post
Yeah but I feel like the whole cozy trend is a backlash to all the gritty and dark media we have had since… late 90s? I remember that the dark tone was something we were craving back then. I wouldn’t be surprised if it has reversed now 30 years later
comment in response to post
I mostly use the vector tools in Figma to tweak icons and since you are working on so small sizes that’s why you get problems. I just realized that maybe I can work on a large scale and the scale it down to avoid the snapping? Will try it
comment in response to post
Figma lacks a lot for vector design since it snaps the endpoints to the grid. This can make the design look wonky. You can use it for layout though and use Inkscape for the vectors
comment in response to post
Cykling. En cykel består av 500 pinaler och varje pinal kan man köpa custom från en liten tillverkare i Sheffield eller Italien eller någon annanstans. Oändligt nörderi
comment in response to post
Even talking is friction, that’s why we have this handy brain chip
comment in response to post
I wrote about this last year but it's an evergreen problem. The backend gets designed according to the logic of tech. The business model gets designed according to the whims of business. These two things then create a huge amount of problems for the actual user, and UX gets stuck making bandaids.
comment in response to post
I think gold rush is the right word. Since, like, 2015 the tech scene has been pretty stable. You, mostly, haven’t been able to strike it rich by solving easy problems in new domains. That’s why people are putting so much hope in every new niche like metaverse, crypto, ai. They want to get in early
comment in response to post
Also, due to their nature the UI will be average. If you have unique use cases or want a unique experience you will need someone to design it
comment in response to post
With the threading layout we could add a control to collapse and hide a thread. Might be my old bbb forum habits that is coming out
comment in response to post
True. What I found interesting about your post is the framing of RTO as retention. Usually we talk about increased control, real estate costs, or layoffs as the reason. All those things drive away employees so this was an interesting pov that nuances the topic
comment in response to post
I start off with asking both Claude and ChatGPT to write a plan for an idea I have. Then i take each of their plans and cross post them to each other for review and improvement. I do this a couple of times before I have a good plan. Then I use cursor to implement that plan step by step