Re:Ask HN: What's your current AI agentic setup?
by hoatrinhdev on Monday November 09, 2026 @10:34AM
(#50427034)
Separate virtual desktop per task is how I work, too. When a run takes a while, do you stay on that task's desktop and wait, or switch to another?
Re:Ask HN: What's your current AI agentic setup?
by AaronAPU on Tuesday November 10, 2026 @04:49PM
(#50446786)
I’m usually only on a particular desktop while code reviewing. When multiple agents are working on different tasks, I just let them work and queue up other work or code review.
Re:Ask HN: What's your current AI agentic setup?
by hoatrinhdev on Wednesday November 11, 2026 @03:37AM
(#50451027)
That makes sense. Review is the part that needs the focus. With several running at once, how do you know one is ready for review: a notification, or do you just check on them between other work?
Most coding via Opus 5.5.
Codex for text and some non-coding tasks.
Separate virtual desktop and clone on dev drive with VScode for each task (much prefer its diff tool and use it to launch builds etc).
I never come close to running out of tokens because I still do code reviews often enough to throttle usage.