Of my friends who use coding agents heavily, the happiest seem to fall into two distant camps:
a) Controlled fast loops: 1-2min cycles, mostly single-threaded, still totally in control of the code, using the agent "to type faster"
b) Delegated slow loops: Nudging along in the background a couple times a day, while something else (design, writing, etc) is their primary focus; paying ~no attention to code; it's fine if agents get stuck for hours
Like many, I've been trying to make some middle ground work: trying to delegate more than the first camp, while giving the work more focus and technical oversight than the second. My role is planning, technical guidance, and code review. This leads to 10-30min cycles, which lead to parallelism (to avoid busy-waiting), which leads to context-switching and fragmentation, which leads to working memory churn and poor comprehension, which leads to situations where neither the agents nor I understand what's going on.
It sucks and I hate it. I'm moving faster, but the work is unpleasant and unrewarding. It seems awfully hard to exert *partial* technical control—much easier to exert ~full or ~none.
The ideal, maybe, is something like what @simonlast outlines in https://t.co/M6R3EF5Yy7: teams of agents making technical plans, reviewing each others' work, autonomously and adversarially testing, etc. You still get robust "technical oversight"—just by other agents. Unfortunately in my domain (mobile interface with heavy gestures and animation) that's not yet tractable, even with lots of homegrown scaffolds and probes. But things will probably look very different in a year.
I'm curious if others have found happy middle ground between these poles?
Neighbors have put mattresses and plexiglass up in their windows to block the noise from this data center in Virginia.
It's a high pitched whine from the natural gas turbines that power it.
The noise never stops 24/7.
One of the best advantages of functional programming is that it throws almost any dev into a pit of success. I don't have to be a hotshot architecht to make things which are robust and easier to maintain.
I just install different kinds of terminals (iterm, wezterm, terminal etc.) and use one for one project. No confusion about which tabs are for which projects. I then put them each in their own desktops so alt-tab also works correctly.
@siddharthkp My theory is that listening to podcast and trying to do the exercise is splitting your attention and probably leaning towards the podcast. If you seriously want to give gym a try, give it your 100% and see if you can feel it in your muscles when you exercise.
So many hard code tech company ceos, like database or network related, are also feeding in AI psychosis because they are planning or selling something to run coding agents on. Everyone is selling something here all the time.
@davidcrawshaw But will those learnings solidify without you getting into the weeds and doing it multiple times or will be similar to how we feel after reading a book?
Something happens to my brain after agentic coding that I can’t describe. It’s like cognitive offloading which folks have already written about, but even more. It feels like I can’t think through problems anymore. Like a fog. Using agentic but losing my hard-won agency.
@VicVijayakumar Reminds me of the joke where a person goes to a doctor and tells him that every part of his body pains. wherever he keeps his thumb and presses.