31–40 of 55 posts
Maybe just me, but I have given up on flow state and my solution is to run 2 or 3 projects concurrently and switch between them. I use virtual desktops. work on one, give an assignment to my AI, switch to the next and cycle through them all day long.
Feels wrong but it's the closest I got.
Bootstrapping things like a new project ? Yeah, things can be too fast for me to stay in the flow.
But my day to day is about working with already existing code which I have to modify and here, AI is exactly the opposite : it helps me drill down boring and legacy code. I helps me stay in the flow because I can ask questions about the code. I can ask it where is the code that does X, I can ask it how it currently works. I can describe my issue, let it analyze the code and ask it to make proposals to solve my issue. Then I discuss the options, then I let it implement the one "we" agreed upon, then I review and I discuss the solution.
In fact I stay in flow because I don't feel alone with my issues and my questions. Maybe it says more about my work environment where I cannot pair as much as I wanted to but at least I have Cursor/Claude/Whatever.
However not having that continuous interaction and hoping the AI to solve your Jira ticket in one prompt is going to be a disaster for your focus and you will not trust the result enough.
Overall, it makes my job less miserable when I'm doing boring things.
If you stop and review what an agent did, it breaks flow.
Personally, my experience has been the best with doing this:
- Start coding
- On a second laptop, run an agent in yolo and tell it to periodically pull changes and if there are any `AGENT-TODO:` items in the code, do them.
- As you code, if you find something irritating or boring, `AGENT-TODO: ...` it.
- Periodically push your changes.
- Periodically pull any changes your agent has pushed down.
- Keep working; don't stop and check on the agent. Don't confirm every action. Just yolo.
If that's too scary, have it put up PRs instead of pushing to the live branch. /shrug
...but, tldr: if you're sitting there watching an agent do things, you're not in flow. If you're kicking multiple agents off and sitting waiting for them, you're more productive, but that is absolutely not flow state.
Anyone who thinks they are, doesn't know what flow state is.
The key to maintaining flow is having a second clone, or a second machine or something where you can keep doing work after you kick the agent off to do something.
(yeah, you don't need a second laptop, but it's nice; agents will often run things to check they work or steal ports or run tests that can screw with you if you're on the same machine)
Yeah, when using AI assistants it often happens that I end up in a dull/low-energy/lethargic state. Then it costs me a lot of energy to get back working on the problem and fix all the mistakes that the AI made.
A hard bug used to be frustrating, but also a bit challenging.
Not anymore, really.
Interestingly I feel quite the opposite but it depends on the job. Bootstrapping things like a new project ? Yeah, things can be too fast for me to stay in the flow. But my day to day is about working with already existing code which I have to modify and here, AI is exactly the opposite : it helps me drill down boring and legacy code. I helps me stay in the flow because I can ask questions about the code. I can ask i…
I think OP is saying it's too slow to stay in the flow. That's what I feel personally. The AI thinks a long time and I do something else.
https://www.tabulamag.com/p/too-fast-to-think-the-hidden-fat...