5 tasks. June 1, 2026 closed at 27.4x weighted leverage across 52.0 human-equivalent hours in 114 minutes of wall-clock time. Supervisory leverage came in at 390.0x.
That is 1.3 weeks of human-equivalent throughput in 1.9 hours. The ceiling was 90.0x; the floor was 6.7x. 4 of the 5 entries came from a single project.
Task Log
| # | Task | Human Est. | Claude | Sup. | Factor |
|---|---|---|---|---|---|
| 1 | ADR 0005: GED novel adult-learner activity set — triaged 16 activities into 6 tiers with [ip] considerations, implementation plan, validation metrics, and risk evaluation; updated ADR index | 15.0h | 10m | 2m | 90.0x |
| 2 | Aegis Phase 1 close-out — scan endpoint + dedup-aware persistence: 5 ORM models (repos/scans/scan_results/findings/finding_events) + Pydantic schemas + orchestrator (sequential scan execution with… | 22.0h | 26m | 1m | 50.8x |
| 3 | Phase 0 of ADR-0005: extended activities catalog with 16 GED novel-activity entries + 4 addons + academic/GED category + deferred runtime_source + per-activity feature-flag plumbing + library mirror sync + 8… | 8.0h | 18m | 1m | 26.7x |
| 4 | Forensic diagnosis of crashed cross-session [content generation] run + safe checkpoint resume (CompTIA Network+) | 2.0h | 15m | 3m | 8.0x |
| 5 | Brought IB Business + AZ-140 to pristine + beta (cohort now 30): generated ~6,157 questions from 0, reweighted both, amplified pairs (IB median 13->15, +313; AZ-140 +34), stamped AZ-140 readiness-gate… | 5.0h | 45m | 1m | 6.7x |
Aggregate Statistics
| Metric | Value |
|---|---|
| Total tasks | 5 |
| Total human-equivalent hours | 52.0 |
| Total Claude minutes | 114 |
| Total supervisory minutes | 8 |
| Total tokens | 510,000 |
| Weighted average leverage factor | 27.4x |
| Weighted average supervisory leverage factor | 390.0x |
| Human-equivalent weeks | 1.3 |
Analysis
The highest factor of the day came in at 90.0x and the lowest at 6.7x, a spread of 13.5 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 15.0 of the 52.0 human-equivalent hours, or 29 percent of the day. No single task dominated the total, so the weighted average is representative.
Supervisory time was 8 minutes against 114 minutes of execution, a ratio of about 1 to 14. Supervisory leverage of 390.0x 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.