Lukas

Fullstack TS developer
Currently in college getting a CS degree

[Lukas @quick007YT]

@okaytanvir imo instant and wipe are the only valid choices (although I'll be honest I think there are better/cleaner wipe animations)

[Lukas @quick007YT]

@Polymarket

[Lukas @quick007YT]

@RhysSullivan realistically you know more about how the bar functions than whoever built this

[Lukas @quick007YT]

@Heistedxyz Redeem

[Lukas @quick007YT]

@voratiq interesting! benchmarking is really hard and seeing more is always nice

[Lukas @quick007YT]

@lindadong @figma does the glow actaully work in figma?

[Lukas @quick007YT]

@ansonlin *isn't

[Lukas @quick007YT]

@ansonlin maybe? but they're unprofitable and I personally (as someone who designs but it's a designer) open figma far less frequently than I used to

[Lukas @quick007YT]

@jxnlco I think just pulled my sora username

[Lukas @quick007YT]

@Antrunt @fofrAI 5.5 is skeptical but sad it can't get this...

[Lukas @quick007YT]

@JustinGorya wait. actually unlimited? and is there a cli or something so I can use it in the codex harness?

[Lukas @quick007YT]

@deepfates "You're right to push back on that"

[Lukas @quick007YT]

@StatisticUrban even microsoft is in awe of how bad twitter managed to make this

[Lukas @quick007YT]

@TheRohanVarma generally just the ergonomics of a gui being far better than a tui. I loved opencode for a while, and I still use it occasionally, but my productivity increased in an app that actually felt like it could keep up with me

[Lukas @quick007YT]

@thsottiaux how do you guys manage right now? I find myself going back to opus or other competent models because it's such a pain to get codex to make something that looks remotly good

[Lukas @quick007YT]

@thsottiaux codex mobile is still quite broken! I was trying to send feedback from my phone but I got an error even trying to do that 🫠

[Lukas @quick007YT]

@DataChaz I would be very skeptical of tools like this. evan already tested this and found increased token consumption, while I would also bet this also makes caching worse

https://t.co/1SF9dmPS4Q

[Evan Boyle @_Evan_Boyle]

I integrated this into Copilot and ran some evals this afternoon. Results were neutral at best and in most cases resulted in MORE token consumption. These results are consistent with our prior evaluation of RTK.

Compression ends up removing information that the agent needs, which triggers a re-read, ultimately resulting in larger costs and latency.

Beware of things that sounds too good to be true.

[Lukas @quick007YT]

@_Evan_Boyle also this one in particular probably fucks up caching too

[Lukas @quick007YT]

@_Evan_Boyle my other concern with a lot of this is agent performance. I can't imagine lobotimizing the agents in the way that many of these types of frameworks do is productive