Building svgapp.ai. Also whatever tool I need while developing.

Joined October 2014
98 Photos and videos
Pinned Tweet
Want AI mascots that stay consistent across poses and scenes? I wrote a practical guide that shows exactly how to do it. ✅ Every step explained 🧰 Tools that improve results a lot 🎨 Works with any image generator svgapp.ai/blog/consistent-ma…
1
4
1,003
I've been using figma to create ad creatives quite a bit recently. It is insane how BAD their AI experience is. I've yet to find a single use case that isn't entirely broken and not doing what it is supposed to do. Especially if you try to create variations of something. Back to making my agent generate them in HTML and then doing a snapshot of that via Javascript as export.
19
I’ve been using a scrolling window manager for some weeks now. Tried to go back to tiling and I gotta admit, i didn’t make it through a full week without switching back. Scrolling wms like niri are just productivity kings
7
First time one of my posts really nailed it. Ranking #1 plus AI overview
14
Happy to announce that svgapp is no longer at 0$ mrr. (Even tho it made profits since launch) After several months of running single purchase I reworked a lot of the app and switched to subscriptions. Within 2 days my SEO based funnel acquired the first subscription based customer 🍀
1
30
Just shipped the probably biggest change to svgapp. You can now collaborate in real time on your creations, share styles with others you’re working with and collaborate as a team on whatever you’re creating. This allows svgapp to no longer only be a product for indie devs but entire teams to quickly ship consistent illustrations and mascots
19
Perfect time for a GitHub alternative startup just make migration easy
all our release PRs, present and past have disappeared @github help?
1
49
No way those tests just finished in 3min 40s on @useblacksmith on first try (so caches weren't even warmed up yet) 💀
Something is up with Github actions last few weeks. The exact same e2e tests running in playwright 2 weeks ago now take up to 3x longer (5min -> 15min runtime). No changes to the test code. It does not run against any remote services, spins up an entire env locally. Why??
71
Something is up with Github actions last few weeks. The exact same e2e tests running in playwright 2 weeks ago now take up to 3x longer (5min -> 15min runtime). No changes to the test code. It does not run against any remote services, spins up an entire env locally. Why??
199
Felix Mennen retweeted
May 14
im not an honest person i squash my commits
62
35
1,321
60,020
rare w commit
24
Lodash was a (bad) solution to a problem due to terrible design choices of JavaScript. It had its place but it’s also good to be forgotten. Also it’s slow and bad and any decent AI can rebuild a 10 line function for whatever you need that runs 200x faster without having to load 80kb deps
anyone remember lodash?
36
Vite react convex stack is just godlike dev ux right now
FINALLY! I've been saying this for a while. Most apps would be just fine using React, Vite and TanStack Router/React Router. Sadly, people love to overcomplicate things.
24
This is the way. Plans are worse than just going at it and steering afterwards / during with a capable model
May 7
i never make plans i hate looking at markdown i don't wanna read markdown files i just plan by having it make changes to the code then i look at the code to see what sucks then i prompt again
21
My openclaw decided to hate me 3 minutes ago
11
I am so happy I have an old af developer account
damn nvm android app is gonna take forever even if youve already converted it i 10000% do NOT have 12 friends with androids lmfaoo
20
Felix Mennen retweeted
May 3
every ai company that has raised a lot of money is probably building a github competitor rn
126
30
2,017
157,229
Cyberpsychosis but instead of your brain handling 10 implants it handles 10 vibe sessions.

ALT Edgerunners Cyberpsychosis GIF

it's so addicting to juggle 10 agents and "make progress" on 10 different apps but never ship anything, this is the true psychosis state
26
So apparently this being in research preview means that users on pro plan have no way of properly testing it since it consumes its entire usage for the setup of a single design system.
Apr 17
Introducing Claude Design by Anthropic Labs: make prototypes, slides, and one-pagers by talking to Claude. Powered by Claude Opus 4.7, our most capable vision model. Available in research preview on the Pro, Max, Team, and Enterprise plans, rolling out throughout the day.
35
I didn’t even get to iteration one with the pro plan fml
yesterday: I gave Claude Code 5 years of design files with Opus 4.7 using Paper Design MCP and it generated hundreds of assets using my coded skills and workflows. 3% of usage today: test Claude Design for hours... 20 iterations into the same design system, rate limited before even finishing. All Anthropic is doing here is throttling our usage with a new “Design” UI and calling it a feature. this on the $200 (20x) Max plan btw...
1
3
208