slack last days..
The only way to kill Slack's new tray icon is to override it from the OS level via macOS "Menu Bar" settings. There is no way to do it from Slack's preferences.
I know at the end of the day everyone is here to collect a paycheck, but I have a difficult time not making a value who implemented this. The banality of evil.
For the good of your users you have to fight this crap, no matter what the PM says.
0
0
0
5
Andrii M retweeted
If I ever get ultra rich, I won’t tell anyone, but there will be signs
The signs:
The signs:
I'm now building an "improvements" feature into all of my personal vibecoded applications: a freeform text input that's hooked up to an agent that yolo implements whatever is submitted.
It's an interesting experience -- the activation energy barrier to sanding any rough edge or ature becomes extremely low. Someone else who uses one of the applications I created has already implemented 150 improvements over the course of a few days. This pattern feels like a new kind of tool -- liquid software? -- and every application that doesn't have this feature now has a slightly rigid and anachronistic flavor. Why can't I just tell it how I want it to be different?
Now, it's not obvious to me how this pattern can or should scale. Does it become unmaintainable after a while? (Maybe model intelligence improvement outstrips my feature appetite such that it's fine?) How does collaborative software work in this model? Can software designed this way be secure and reliable?
Whatever the larger questions about broader deployment, it definitely works great in a personal context. It reminds me a little of the visions of Lisp/Lisp machines, Smalltalk, and HyperCard -- systems that the user can take control of and modify. Those visions never quite worked, though: the vast majority of software today is not end-user modifiable in any useful way. So, while this "liquid" affordance is a very small touch, the way it changes one's relation to the software feels a bit transgressive and empowering in a Lickliderian and Engelbartian way.
0
1
14
3,447
Andrii M retweeted
If there’s one thing to learn from llms: fail fast and move on.
llms make a bazillion mistakes but also course correct on a dime and don’t keep endlessly bringing up the failures.
Look how far it’s gotten them.
llms make a bazillion mistakes but also course correct on a dime and don’t keep endlessly bringing up the failures.
Look how far it’s gotten them.
15
4
204
5,890
this is actually a pretty cool idea of mcp feedback 🗒️📓📝
docs.sentry.io/product/user-f…
We are also experimenting in the MCP with having agents automatically send feedback when something didn’t work the way they expected (using this same API)
0
0
0
23
this is a great lesson how to take the feedback even if it is negative on X.
@zebassembly @thosiawa @paper Really good feedback and I agree, we’ll look into making that smoother
0
0
1
98
thanks @thsottiaux I don’t even need any new terminal, just “codex laptop” with YouTube and music options. that is it! (no touch screen pls)
Day 26/ You can now easily switch between multiple accounts on mobile, and coming soon to desktop too. Next two days will be resets… banked… do with this what you may 👀.
help.openai.com/en/articles/20…
0
0
1
76
Andrii M retweeted
You can build a software business without AI
Or you can build one for $200/mo
Or you can burn a bunch of investor capital and build one
And almost all of those scenarios go to zero but the ones where you spend only a sustainable amount of capital are better odds.
Software has always been easy to build people need to stop convincing themselves otherwise.
Or you can build one for $200/mo
Or you can burn a bunch of investor capital and build one
And almost all of those scenarios go to zero but the ones where you spend only a sustainable amount of capital are better odds.
Software has always been easy to build people need to stop convincing themselves otherwise.
What we are seeing at @coderabbitai and elsewhere on this parabolic trajectory is that even larger seed rounds ($8M+) are no longer enough to build a software business. Software factories need quality tokens and large-scale compute to run CI, sandboxes, and other infrastructure to ship software at scale. At this point, smaller startups exist only to funnel ideas to larger startups with more tokens.
31
11
317
21,696
Andrii M retweeted
The coding interview is dead, at least for my team at Anthropic. I've seen enough to know that it no longer provides signal. I removed it from loops this month.
I have a spicy, optimistic take - I think there's an opportunity for junior engineers: felixrieseberg.com/hunger-beats-t…
I have a spicy, optimistic take - I think there's an opportunity for junior engineers: felixrieseberg.com/hunger-beats-t…
630
1,497
15,016
2,662,276
