10 tasks. May 8, 2026 closed at 22.4x weighted leverage across 108.5 human-equivalent hours in 291 minutes of wall-clock time. Supervisory leverage came in at 323.9x.
That is 2.7 weeks of human-equivalent throughput in 4.8 hours. The ceiling was 53.3x; the floor was 4.7x. 10 of the 10 entries came from a single project.
Task Log
| # | Task | Human Est. | Claude | Sup. | Factor |
|---|---|---|---|---|---|
| 1 | Browse-before-auth web client implementation: all 5 phases (router public/gated split, pendingIntent + resumeAfterAuth + AuthCallback dispatcher, anonymous CourseDetail with auth-aware Enroll, AppShell… | 40.0h | 45m | 1m | 53.3x |
| 2 | ADR-0003 Phase 1 engine: PosteriorWarmStarter module, BetaPosterior trust_flagged field, mastery trust gate, CreateAutopilotRequest/Response field expansion, 5 CrossDomainConfig fields, cloud.toml section,… | 7.0h | 16m | 0m | 26.2x |
| 3 | Pair-to-node ref repair across 247 broken domains via embedding cosine match (146,762 pairs re-anchored, mean cosine 0.91). Bulk readiness-gate stamp across 178 manifests derived from exam metadata.… | 12.0h | 30m | 1m | 24.0x |
| 4 | Documentation | 7.0h | 18m | 2m | 23.3x |
| 5 | ADR-0003 Phase 2 client wiring (web + Electron): API types, env flag, autopilot store extensions, CrossDomain fast-track buttons, CourseDetail savings callouts, SkillsCarryingOverPanel warm-start data, i18n… | 5.0h | 14m | 0m | 21.4x |
| 6 | iOS cross-domain fast-track parity (EngineClient types, AppState TransferContext, CrossDomainView fast-track button, AutopilotView pre/post-activation callouts, env flag), invite-code gate removal… | 5.5h | 18m | 0m | 18.3x |
| 7 | Domain pair-to-node integrity audit (323 domains, 76% degraded), EB leaf catastrophic-regression fix (gate on domain_obs_total instead of raw pair_stats — acc92 crashed 1.0→0.001 on broken-pair domains),… | 18.0h | 65m | 4m | 16.6x |
| 8 | ADR-0003 Phase 3 artifacts: 5 reference profile YAMLs (CLF→SAA, SAA→SAP, AZ900→AZ104, CKA→CKAD, CISSP→CISM), run_warmstart_validation.py synthetic A/B harness (~500 lines, parses clean),… | 4.0h | 15m | 0m | 16.0x |
| 9 | Built shared [engine subsystem] server (FastAPI/MPS) + [model] embeddings client + engine wiring so [engine subsystem] can run 10-way concurrent without OOM | 6.5h | 25m | 4m | 15.6x |
| 10 | Resume ADR-0003 cross-domain warmstart work after crash: bundle drift into 4 focused engine commits + 1 web a11y commit, add Phase 11 to avian-audits content audit (md spec + py implementation) catching… | 3.5h | 45m | 7m | 4.7x |
Aggregate Statistics
| Metric | Value |
|---|---|
| Total tasks | 10 |
| Total human-equivalent hours | 108.5 |
| Total Claude minutes | 291 |
| Total supervisory minutes | 20 |
| Total tokens | 1,425,000 |
| Weighted average leverage factor | 22.4x |
| Weighted average supervisory leverage factor | 323.9x |
| Human-equivalent weeks | 2.7 |
Analysis
The highest factor of the day came in at 53.3x and the lowest at 4.7x, a spread of 11.4 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 40.0 of the 108.5 human-equivalent hours, or 37 percent of the day. No single task dominated the total, so the weighted average is representative.
Supervisory time was 20 minutes against 291 minutes of execution, a ratio of about 1 to 14. Supervisory leverage of 323.9x 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.