Comment by bicx
7 days ago
For me, flow state can be achieved through the true multitasking that agentic coding allows. Start an agent on a task, then set up another agent for another task. While one is working, the other usually needs feedback or tweaking. I run Claude Code in yolo mode in a dev container where it can cause little damage, so I’m not wasting my time approving tool calls (or managing an allowlist of tool calls). Sometimes I have 3 agents going at once, but I’ve found that 2 is usually as many as I can manage. I suppose if your project specifications are really dialed in, you could have more than that.
This is my flow as well. Though I have to say, on days where I drive this to the extreme, I can feel the tokens oozing out of my brain in the evening, it's definitely its own version of draining.
Is isolating files they work on an issue for you? Does this simply not pop up (because you avoid it by having a good understanding of what will be touched) or are you doing a lot of merging?
Git worktrees are one way of doing this, another is to have the agents work on separate components (backend, frontend, etc.) or different tasks alltogether (review, ideation, documentation, etc.)