Seriously, what do you do while Claude (or whatever you use) is doing its thing?
10 Likes
The main problem to me is that you do not always know if you are going to wait for 1 minute or 10. That can be very annoying. On good sessions though the back and forth is relatively stable, but too short to really invest your brain in something else and too slow to just stare at your screen.
And context switching is tiring, so if I can I write specs for the next prompt / next task, otherwise I just surf the web. You can also review the code as it is edited to save time.
3 Likes
I carefully look over her shoulder to read all the thinking outcome. To ①know what the heck do we do, ② be able to interrupt her at any time she’s driving bonkers, and ③ make it easier to CR after everything is done.
9 Likes
Huh? It’s extremely dangerous to give an LLM tasks you could not accomplish yourself (because you won’t be able to validate the result.) But if you can write the code you probably should know how much time would it take.
So yeah, I always know how long am I to wait (and it’s never 10 mins, there is next to zero chance LLM is able to accomplish such a long task without hidden uncertainties, which will surely shoot her in the foot later.)
3 Likes
No it’s not dangerous if you know what you are doing. It has to read all the specs I wrote, understand the full codebase to make the right changes.
Review is not a problem as I know the codebase very well. (I wrote 80% of it, if not 98% depending on projects).
Admitedly my worse timings were on slow 4g network but still, Claude with high thinking, peak usage and a large codebase can be so slow.
But if you can write the code you probably should know how much time would it take.
I know how much time I would take to write it, yes.
5 Likes
I’ve always said that the basic concept of algebra (and indeed of all mathematics) is proportion!
This looks like a nice saying yes, but its not worth much more in this context.
It’s very obvious to me that when you open a codebase (or just started claude) you will take a lot of time to do the most basic feature, whereas when you know the codebase (or claude has many relevant files in context) you will take much less time to do something that is actually more complex than the first task.
Sometimes I look at my side projects for any small wins, other times I catch up on slack. Mostly talk to friend on WhatsApp (this gives me extra energy to continue with my work day at the cost of my wife thinking I dont work at all.
3 Likes
I had one project this week that Claude, on Sonnet 4.6 high, took 4 days to complete - with no input after the initial prompt. I was surprised it took that long but it did. Not a general purpose project by any means, and it had a formal spec, a conformance test suite and several reference implementations to work from so it really didn’t need me in the loop.
What do I do when it’s doing its thing? Have a few more sessions doing other things!
Or, increasingly, go outside and enjoy the fresh air. And construct something physical, not virtual 
8 Likes
I tend to do a combination of [1] follow along with it (though it can start to feel crazy after some time because of how quickly it goes), [2] work on other parts of the project, [3] research or do external marketing work, and [4] step away from the computer to be present in other areas of my life.
I’ve noticed the most gains in quality from Opus 4.6 (and 4.7 is a bit shaky at times —> either being spot on or quite sloppy, and I did read that you need to prompt it differently than you would for say 4.6). In Tidewave, I’ve tended toward using 4.7 (except for yesterday it seemed to behave strangely) for fixing specific a11y compliance and 4.6 for less specific coding problems (like working on new features or reviewing things). So, despite the cost, Opus 4.6 has been a positive shift in working with an LLM (kind of like having someone to do the day laboring for you while you architect/manage the work).
1 Like
I eat junk food and give my brain calories to fuel up for the next round. I find claude coding sessions to be cognitively intense! For me it will take awhile to build optimal work habits.
4 Likes
More sessions…?
Usually two or three with stepwise milestone plans or smaller single steps. So when one works I check on the output of what another session has done since the last step and test or check on it. Then write a new prompt for it. Depending on the project this keeps me busy, and Claude has to wait for me.
The notable exception being some very data intensive ones were it can take a very long time for each step. If I know it will take an hour or three I just do some other kind of work entirely, or try do it in the background while I do research and planning.
I also have this ongoing skill development, so I will ask the sessions to give feedback of the skills used during or after use, and then spend time trying to evaluate and make improvements accordingly.
Sometimes I roam around this forum
4 Likes
This is actually one of the more interesting threads on LLMs I’ve read in a while.
My experience so far as been far has been working far more exploratory and chaotic. Most of my prompts are very precise and usually don’t take very long at all.I’ve only had a couple of times where something has taken… I’m not even sure, maybe 30 minutes? But after reading Kip’s answer I want to try something like that!
In any event, I’m usually thinking, “Is this really what I want to do?” and then when it’s done I say, “Wait, no, let’s try implementing it this way” and it says, “Yes, you are very smart, that is a much better idea!” and then when it’s done I say, “Oh wait no, I tried the new way and the old way is actually better, let’s revert back to it” to which it says, “Of course! You are very smart and the old way so obviously better, I should have seen that!” I keep forgetting to enable the Caveman plugin.
Otherwise, my answer is pretty simple: YouTube, which still includes videos of people talking about how much they hate AI even though I’m fully bought in at this point 
3 Likes
Usually will prepare the further instruction to be typed. And review the code generated from Claude Code last session.
Sometime working on frontend code, will do some regression testing manually.
2 Likes
As of a few months ago… pushups
I’m up to 78 in one go
10 Likes
That’s awesome!
In the past week, I have had a couple tasks where the agent is “thinking” for 5 solid minutes on a single reply prompt. No back-and-forth with tool calls, like one single reply. I even thought that one of them timed out, so I cancelled it and re-ran the query, and the response took another 5 minutes. This was with Claude Opus 4.6.
2 Likes
I have a rule “Watch the prompt response and treat no outcome for more than 30 secs as a timeout, report it and ask what to do next.”
The 5-mins thinking is a definite hint that the outcome will be not ideal, to put it gently. It’s like a junior developer constantly saying at standups “I have some progress,” meaning there is no progress whatsoever.
1 Like
