8 tasks. March 29, 2026 closed at 20.8x weighted leverage across 226.0 human-equivalent hours in 652 minutes of wall-clock time. Supervisory leverage came in at 237.9x.
That is 5.7 weeks of human-equivalent throughput in 10.9 hours. The ceiling was 96.0x; the floor was 5.3x. 8 of the 8 entries came from a single project.
Task Log
| # | Task | Human Est. | Claude | Sup. | Factor |
|---|---|---|---|---|---|
| 1 | Full readiness audit: 39 repos, ~195 checks, 4218 tests, 7 MEDIUM + 13 LOW findings, 10 commits pushed | 24.0h | 15m | 3m | 96.0x |
| 2 | Add topics_and_objectives and exam URLs to 658 domain specs | 6.0h | 8m | 3m | 45.0x |
| 3 | Deterministic diagram audit script + fix 46 diagram issues across N-Y | 40.0h | 55m | 5m | 43.6x |
| 4 | [ip] diagram quality overhaul: collision detection + overlap detection + edge straightening + 74 arrow style fixes + 23 numeral fixes + spec text fixes + ROUTING-ALGORITHM.md | 120.0h | 360m | 30m | 20.0x |
| 5 | Cross-check filed [ip] diagrams A-H against specifications: 61 diagrams vs 8 specs auditing reference numerals and figure titles | 8.0h | 25m | 5m | 19.2x |
| 6 | Full deployment readiness audit — 42 repos, 351 checks, 4321 tests, 8 issues fixed | 8.0h | 33m | 3m | 14.5x |
| 7 | Full deployment readiness audit: 42 repos, 417 checks, 4321 tests, security fixes, SEO fixes, doc fixes | 16.0h | 111m | 5m | 8.6x |
| 8 | Full [ip] portfolio audit (7 phases across 17 CIP apps + supporting docs) | 4.0h | 45m | 3m | 5.3x |
Aggregate Statistics
| Metric | Value |
|---|---|
| Total tasks | 8 |
| Total human-equivalent hours | 226.0 |
| Total Claude minutes | 652 |
| Total supervisory minutes | 57 |
| Total tokens | 2,274,500 |
| Weighted average leverage factor | 20.8x |
| Weighted average supervisory leverage factor | 237.9x |
| Human-equivalent weeks | 5.7 |
Analysis
The highest factor of the day came in at 96.0x and the lowest at 5.3x, a spread of 18.0 times between the two. That is a wide range for a single day, and it usually means the day mixed mechanical work with work that needed real judgement.
The largest single entry accounted for 24.0 of the 226.0 human-equivalent hours, or 11 percent of the day. No single task dominated the total, so the weighted average is representative.
Supervisory time was 57 minutes against 652 minutes of execution, a ratio of about 1 to 11. Supervisory leverage of 237.9x is the figure I find most honest, because it measures the hours I actually spent rather than the hours a machine spent on my behalf.
Every figure here is recorded at the time the work is done rather than reconstructed afterwards. The human estimate is my own judgement and carries the uncertainty that implies; the minutes and tokens are measured. The full dataset, including this day, is available for download.