Last week I had two problems:
1) my AI coding costs were random and stressful
2) every “quick break” became 40 minutes of doomscrolling
I’m a solo Mac builder, so both problems hit hard.
Here’s what actually worked for me in 7 days.
1) I stopped guessing token spend
I used to think: “I’m on a plan, I’m probably fine.”
Then I looked closely and realized I was wasting money on:
- repeated context resets
- re-running similar prompts
- tired late-night retries
So I put cost visibility where I can’t ignore it: the menu bar.
I built TokenBar (tokenbar.site), a tiny Mac app that shows live token usage + spend while I code.
Big difference:
- I catch runaway sessions faster
- I tighten prompts earlier
- I stop “one more retry” loops sooner
2) I treated feeds like a trigger, not “entertainment”
My worst cost spikes happened after social feed breaks.
Feed -> context switch -> distracted return -> sloppy prompt -> expensive rerun.
So I blocked feeds during deep work blocks.
I built Monk Mode (mac.monk-mode.lifestyle), a Mac app that blocks distracting feeds at the feed level.
Not the whole internet. Just the stuff that hijacks momentum.
That one change made my coding sessions cleaner and shorter.
3) I started using a simple “focus cost loop”
Before session:
- pick one task
- define done condition
During session:
- monitor token burn live
- no feed apps
After session:
- log what caused biggest spend
- fix that in next prompt pattern
This loop sounds boring, but it compounds fast.
4) Results after one week
- fewer zombie coding sessions
- more predictable daily spend
- less context thrash
- more shipped work
No magic model switch. Just visibility + environment control.
5) If you’re solo-building, start here
If your AI bill feels random, don’t optimize everything at once.
Do this first:
- make spend visible while you work
- remove feed triggers during build blocks
That combo changed more for me than any prompt template ever did.
If you’re also building with Claude/Cursor/Codex and want, I can share the exact daily checklist I use to keep both cost and attention under control.
Top comments (0)