Skip to main content
AI16 April 2026

Mostly content production

5 tasks. April 16, 2026 closed at 46.3x weighted leverage across 112.0 human-equivalent hours in 145 minutes of wall-clock time. Supervisory leverage came in at 320.0x.

That is 2.8 weeks of human-equivalent throughput in 2.4 hours. The ceiling was 87.3x; the floor was 19.2x. 2 of the 5 entries came from a single project.

Task Log

#TaskHuman Est.ClaudeSup.Factor
1Wire entire avian-app-web to engine API: 19 pages (Analytics, Governance, CrossDomain, KnowledgeMap, Trajectory, Readiness, SessionPreview, MicroChallenge, Scenario, CourseDetail, InitializationFlow,…80.0h55m3m87.3x
2Redesign auth-service sign-in view to match accelastudy.ai theme: switch font to Plus Jakarta Sans, replace cyan accent with teal, edge-to-edge background, move footer inside shell, animate 4 independent mesh…6.0h12m3m30.0x
3Debugging12.0h35m8m20.6x
4Add 5 missing MCP tools (verify_mfa, get_subscription, get_payment, reactivate_subscription, simulate_webhook payment_id), update cancel_subscription_for_user with at_period_end, wrap MCP responses, fix all 6…6.0h18m3m20.0x
5Audit and review8.0h25m4m19.2x

Aggregate Statistics

MetricValue
Total tasks5
Total human-equivalent hours112.0
Total Claude minutes145
Total supervisory minutes21
Total tokens1,080,000
Weighted average leverage factor46.3x
Weighted average supervisory leverage factor320.0x
Human-equivalent weeks2.8

Analysis

The highest factor of the day came in at 87.3x and the lowest at 19.2x, a spread of 4.5 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 80.0 of the 112.0 human-equivalent hours, or 71 percent of the day. One task carrying that much of the total is worth noting; the average is doing less work than it appears to.

Supervisory time was 21 minutes against 145 minutes of execution, a ratio of about 1 to 7. Supervisory leverage of 320.0x 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.