11 tasks. March 6, 2026 closed at 113.9x weighted leverage across 374.0 human-equivalent hours in 197 minutes of wall-clock time. Supervisory leverage came in at 431.5x.
That is 9.3 weeks of human-equivalent throughput in 3.3 hours. The ceiling was 288.0x; the floor was 30.0x. 6 of the 11 entries came from a single project.
Task Log
| # | Task | Human Est. | Claude | Sup. | Factor |
|---|---|---|---|---|---|
| 1 | Build AccelaStudy React web app - full study activity suite with auth/routing/stores/tests | 120.0h | 25m | 5m | 288.0x |
| 2 | Infrastructure | 120.0h | 25m | 8m | 288.0x |
| 3 | Create accelastudy-infinite-web repo with reverse-engineered requirements and technical design docs | 16.0h | 12m | 5m | 80.0x |
| 4 | Build predictive leverage scoring with Anthropic Claude API integration | 16.0h | 12m | 5m | 80.0x |
| 5 | Documentation | 24.0h | 25m | 5m | 57.6x |
| 6 | Comprehensive acquisition scenario analysis document (609 lines: 7 acquirer categories 3 timing scenarios deal structures founder outcomes 15 comparable transactions) | 24.0h | 25m | 5m | 57.6x |
| 7 | Build MC question bank generator with knowledge + scenario question types and run CLF-C02 | 40.0h | 45m | 5m | 53.3x |
| 8 | Coding | 4.0h | 8m | 5m | 30.0x |
| 9 | Add Cross-Domain Semantic Intelligence Engine (Subsystem 3.25) to AVIAN Reference Architecture | 2.0h | 4m | 3m | 30.0x |
| 10 | Add NDA-gated auth to avian-demo (login screen + CAPTCHA + user list + dev bypass) | 4.0h | 8m | 3m | 30.0x |
| 11 | Update Origin Reference Architecture with S900 Cross-Domain Intelligence Engine subsystem | 4.0h | 8m | 3m | 30.0x |
Aggregate Statistics
| Metric | Value |
|---|---|
| Total tasks | 11 |
| Total human-equivalent hours | 374.0 |
| Total Claude minutes | 197 |
| Total supervisory minutes | 52 |
| Total tokens | 1,048,000 |
| Weighted average leverage factor | 113.9x |
| Weighted average supervisory leverage factor | 431.5x |
| Human-equivalent weeks | 9.3 |
Analysis
The highest factor of the day came in at 288.0x and the lowest at 30.0x, a spread of 9.6 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 120.0 of the 374.0 human-equivalent hours, or 32 percent of the day. No single task dominated the total, so the weighted average is representative.
Supervisory time was 52 minutes against 197 minutes of execution, a ratio of about 1 to 4. Supervisory leverage of 431.5x 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.