Live data from Hacker News

Ask HN: How do you get into a flow state when using AI to code?

news.ycombinator.com

131–140 of 140 posts

Re: Ask HN: How do you get into a flow state when using AI to code?

#132

You can't, "agents" and "agentic coding" are incompatible with the deep understanding of your code that getting into the flow state requires. By definition, prompting a tool to do your job for you means giving up the depth of control that comes with actually writing the code. Some people can work like that, some very few people may even produce acceptable results like that. I cannot and I am not interested in doing s…

Also just having the ease of access to an LLM means you're more likely to reach for it. Combined with intermittent rewards, it's mentally exhausting.

I’m convinced that the intermittent rewards and the way the rewards are variable directly replicates the most addictive parts of gambling.

Re: Ask HN: How do you get into a flow state when using AI to code?

#134

Read the blog posts from Peter Steinberget blog steipete.me about his setup. Many AI builders converged on using the terminal with multiple panes. When a prompt is running if you don't have anything to do add another pane and start another prompt. Little by little you might have 1-6 prompts in parallel at some point. The flow state with AI is managing productively several prompts in parallel.

This is multi-tasking and the opposite of "deep focus".

I find it works as focus when the tasks are all related/in the same repo. Not the same as single-task flow state of old style coding, but also still a kind of flow

Re: Ask HN: How do you get into a flow state when using AI to code?

#137
I have gotten into spec driven development. Plan everything ahead of time. Sometimes I have the LLM help out, sometimes I do it all myself. It can include data models, business logic, or UI design. Sometimes I draw on a whiteboard, or in a notebook, and ask the LLM to convert the drawing into text I can put into a SPEC.md file. The more time I spend planning, the longer the LLM can work autonomously, and the less time I spend debugging issues or asking it to fix things that I didn't ask for.
Post reply on HN