Your AI coding, measured.
Sample data Try it yourself
Loading…
Where you stand
Daily rhythm
Every day in range. Click a day to filter the page to it; click again to clear.
The long view
Every day since collection began, one square each. Click a week to filter to it.
Where the effort goes
Repository first, then the folders inside. Click a bar to filter.
By repository
Inside
Model mix
Output by model
Turns by reasoning effort
When you work, and what it costs
Your week against the meter
Your own hours, with every vendor's peak window drawn over them.
Peak and off-peaki
Are you getting better at this?
7-day rolling mean across the whole range.
Tools, delegation and the heaviest sessions
Tool calls
Named subagents
Heaviest sessions
Writes/reads is the produce-to-explore ratio. Peak context is the largest single-turn context.
What to change
Ordered by severity, then by what it is worth. Computed over the whole window — the filters above do not change it.
Findings below are generated in English.
66.9% of your time-priced spend (deepseek, zhipu) ran inside a peak window, where the same tokens cost up to twice the off-peak rate. That timing cost about $39.39 over the period.
- Queue unattended work for an off-peak hour — test generation, migrations, doc sweeps, bulk refactors.
- Leave interactive work where it is: the premium buys your attention, and a batch job does not need it.
- The windows are narrow — DeepSeek 01:00-04:00 and 06:00-10:00 UTC, GLM 14:00-18:00 UTC+8 on weekdays, so weekends are free of it entirely.
116 session(s) peaked above 300,000 tokens on a single turn (worst: 942,469), and every later turn re-reads it.
- Finish the thread and start the next piece of work in a fresh session.
- Carry forward only what that work needs — an exploration transcript dragged into unrelated work is what makes context this expensive.
7 session(s) wrote 933,593 tokens to cache and never read one back — the 1.25x write premium paid for nothing.
Ask the follow-ups in the same session — the second turn is where caching starts paying. These were closed right after the first big context load.
1 session(s) made 7 read-type tool calls with no edit or write, at 0.67 USD. Some is real research; some is hunting for what a targeted search would have found.
- Delegate 'where is X' to a search subagent — the answer comes back without the file bodies.
- Ask the question you want answered rather than reading toward it.
89.1% of everything the models read came from cache rather than being re-billed at full input rate.
No action. Watch this number — a fall means sessions are being restarted more often.
13.1% of output came from subagents — enough to keep bulk reading out of the main context without runaway fan-out.
No action.
checkout-service took 34.7% of estimated spend across 81 sessions; the top three take 82.1%.
Compare this ranking against where you would say your priorities are. A workspace high on this list and low on your own priority list is the finding.
Peak hour is 03:00 (379 turns). 42.6% of turns fall outside 06:00-22:00.
Useful as a check on whether AI use is displacing focus time or filling gaps. A high out-of-hours share is worth knowing about for its own sake.
deepseek is your largest at 36.9% of output tokens, across 5 vendors in total.
No action. A mix is what lets you re-price or re-route without re-learning a workflow.