I had Matt Damon in my head while reading #ProjectHailMary but Ryan Gosling is kinda refreshing www.youtube.com/watch?v=m08T...
30.06.2025 13:25 โ ๐ 3 ๐ 0 ๐ฌ 0 ๐ 0@pascalwelsch.com.bsky.social
Flutter & Dart GDE ๐ Follow me for updates on #Flutter and #Dart Co-Founder wiredash.com ๐ฌ๐ CTO phntm.xyz ๐ผ he/him ๐ฉโโค๏ธโ๐จ husband ๐ฉ๐ช Cologne
I had Matt Damon in my head while reading #ProjectHailMary but Ryan Gosling is kinda refreshing www.youtube.com/watch?v=m08T...
30.06.2025 13:25 โ ๐ 3 ๐ 0 ๐ฌ 0 ๐ 0That's a fantastic ice breaker question for any party
27.06.2025 13:04 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0You can press D and W in Google Calendar to quickly switch between Day and Week view ๐
27.06.2025 11:02 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0I thought JSON is way older ๐คฏ
26.06.2025 18:02 โ ๐ 3 ๐ 0 ๐ฌ 2 ๐ 0They implemented theming very well. I use the same concept for all my internal UI libraries
23.06.2025 18:49 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0taskmaster does an amazing job estimating issues mcpmarket.com/server/task-...
23.06.2025 14:30 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0Isn't Claude perfect for estimating story points? I honestly think humans can't do it better.
In the end, story points are a tool for PM/POs. They now have a tool to get unbiased estimates without interrupting developers. Perfect ๐
Mein talk vom Flutter Day 2024 รผber Widget Testing mit Spot und Testing Robots ist online ๐
www.youtube.com/watch?v=AV21...
Go outside, touch some grass #outage
12.06.2025 18:24 โ ๐ 4 ๐ 0 ๐ฌ 4 ๐ 0Perfect start ๐ #MarioKartWorld #Switch2
05.06.2025 11:04 โ ๐ 5 ๐ 0 ๐ฌ 0 ๐ 0I sometimes prefer building up a mutable list. It might not be the shortest, but is definitly the easiest to debug.
31.05.2025 00:51 โ ๐ 3 ๐ 1 ๐ฌ 0 ๐ 0yup, WTH
30.05.2025 16:07 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0Dart developers! Quick, what's the output?
30.05.2025 14:32 โ ๐ 1 ๐ 0 ๐ฌ 2 ๐ 0spot: 0.18 is a huge release ๐
#Flutter widget testing becomes better with every update
pub.dev/packages/spo...
Engagement is down on bsky, but what about your traffic @codewithandrea.com?
27.05.2025 12:14 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0Try it: Search for "sweet-pascal" in the Extensions view or install it from Visual Studio Marketplace marketplace.cursorapi.com/items?itemNa...
14.05.2025 02:21 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0Everything is open source github.com/passsy/sweet...
14.05.2025 02:21 โ ๐ 3 ๐ 0 ๐ฌ 1 ๐ 0Since I spend more and more time in cursor, I ported my personal editor theme (Sweet Pascal) from IntelliJ to VScode.
The pastel colors are inspired by Sweet Sweet Canyon from Mario Kart 8
Liebes @zdfde.bsky.social, arbeitet ihr schon an Folge 6 oder Staffel 2 von Inside CDU? www.zdf.de/dokus/inside...
06.05.2025 09:01 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0I use claude-3.7-sonnet most of the time
25.04.2025 21:36 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0It's easier if there are exising rules. This worked for me in a new project
25.04.2025 21:04 โ ๐ 2 ๐ 0 ๐ฌ 2 ๐ 0You need to understand that most code is average code. Thus, most training data is average code. Thus, LLMs produce average code.
They know how to write good code but need to be told what good code looks like. Good is very personal, therefore you need personal rules
Whenever I encountered anything I don't like I created a cursor rule. Well, I let cursor write the cursor rule ๐.
Currently I have over 50, ~30 are Dart specific.
I found Steve who more or less pretty much works like me www.youtube.com/watch?v=uJim...
24.04.2025 15:19 โ ๐ 3 ๐ 0 ๐ฌ 0 ๐ 0To me, that's the easier part. Doing animation and other complex UI stuff is where I struggle the most. Also that is very hard to test
24.04.2025 13:23 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0You have to tell it to run the tests and they have to be fast to be useful. But then it quickly iterates towards a valid solution.
What other guardrails do you know to keep any contributor (human or AI) from breaking existing functionality?
Currently, I let cursor write ~80% of my code. With a good set of rules it produces code as I would write it myself, just faster.
24.04.2025 12:38 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 1You can and you should. Read on permutation testing pitest.org
Remove a line of code. When no test fails you either miss a test or the line was unnecessary in the first place.