Skip to main content
AI24 March 2026

Coding and refactoring

39 tasks. March 24, 2026 closed at 88.8x weighted leverage across 1,167.5 human-equivalent hours in 789 minutes of wall-clock time. Supervisory leverage came in at 432.4x.

That is 29.2 weeks of human-equivalent throughput in 13.2 hours. The ceiling was 274.3x; the floor was 15.0x. 31 of the 39 entries came from a single project.

Task Log

#TaskHuman Est.ClaudeSup.Factor
1Implement 9 new [ip] applications (D F G L O R T U V) — 180 claims across 10789 LOC in 18 new files160.0h35m5m274.3x
2Port 10 [ip] features web→electron (full IPC pipeline + 29 files)120.0h35m5m205.7x
3Sync 10 [ip] features to electron (29 files 7031 LOC full IPC pipeline) and iOS (8 views 15 endpoints)160.0h60m5m160.0x
4Draft [ip] applications S and T (Content Staleness + Assessment Composition)40.0h18m5m133.3x
5Implement 3 [ip] applications (R/T/V) with 60 claims as working code40.0h18m5m133.3x
6Implement 3 AVIAN [ip] applications (F G O) - bootstrap+versioning+readiness40.0h18m5m133.3x
7Coding40.0h20m5m120.0x
8Update [ip-research] docs for 25 apps and regenerate 362 PDFs50.0h25m3m120.0x
9Coding24.0h12m3m120.0x
10Build AVIAN static website with 7 pages particle hero scroll animations24.0h12m3m120.0x
11Resequence [ip] portfolio A-Y (25 apps) across 6 repos64.0h35m5m109.7x
12Build all 10 Tier 1+2 [ip] client features (init+scenario+analytics+governance+curriculum+trajectory+multimodal+crossdomain+chat+cognitive)80.0h45m5m106.7x
13Draft [ip] applications U (Explainability) and V (Cohort Intelligence) with 20 claims and 8 figures each40.0h25m5m96.0x
14Sync 10 [ip] features from web to iOS as SwiftUI views40.0h25m5m96.0x
15Build Tier 2 features #6-8: Trajectory/Evidence/CrossDomain UI in avian-app-web16.0h12m5m80.0x
16Build Tier 1 client features: Entity Initialization + Scenario Assessment + Analytics Panel for avian-app-web24.0h18m5m80.0x
17Build Governance Dashboard (App L) + Enhanced Curriculum (App I) screens with full CSS light/dark mode16.0h15m3m64.0x
18Coding8.0h8m3m60.0x
19Write unit tests for 3 AVIAN subsystems (Assessment Composer + Interview Engine + Cohort Intelligence) - 89 tests total8.0h8m3m60.0x
20Flesh out 3 thin API modules with 16 [ip]-aligned public methods8.0h8m3m60.0x
21Fix 24 failing + 2 skipped unit tests across 6 test files in avian-engine8.0h10m3m48.0x
22Documentation6.0h8m3m45.0x
23Coding12.0h18m5m40.0x
24Write unit tests for 3 AVIAN subsystems (bootstrap/versioning/readiness_prediction) - 123 tests8.0h12m3m40.0x
25Deploy issue-tracker to production (ECR+CloudWatch+S3+ALB+CloudFront+Route53+RDS+Docker+frontend+leverage-tracker cleanup)16.0h25m5m38.4x
26Add bug report proxy endpoint to engine and 7 client apps16.0h25m5m38.4x
27Deploy marketing-tracker to production on AWS (ECR + CloudFront + ALB + RDS + S3 + Route53 + Docker)8.0h15m5m32.0x
28Build Tier 2 features #9-10: context-aware chat (App O) + cognitive state detection (App G)6.0h12m5m30.0x
29[ip-research] hardening analysis for 15 CIP [ip] applications (A-P excluding G)16.0h35m5m27.4x
30Implement 4 QC fixes in AVIAN [engine subsystem] ([quality gate] + readiness convergence + pair regeneration + pair validation)8.0h18m5m26.7x
31Coding8.0h18m5m26.7x
32Coding8.0h18m3m26.7x
33Unit tests for 3 AVIAN subsystems: staleness + adversarial + explainability (112 tests)8.0h18m5m26.7x
34Research FreshBooks and QuickBooks and write comprehensive accounting app prompt8.0h20m3m24.0x
35Write targeted unit tests for 3 packages (observation/governance/reembed) to push above 70% coverage3.0h8m3m22.5x
36Deploy marketing-tracker + CodePipeline for both tools + EC2 private subnet migration (NAT gateway + route tables + instance replacement + ALB re-registration)20.0h55m5m21.8x
37Debugging4.0h12m3m20.0x
38Security quick wins: timing attack + JWT audience + Docker non-root for marketing-tracker0.5h2m2m15.0x
39Create ecosystem inventory document for AVIAN docs - 37 repos catalogued2.0h8m3m15.0x

Aggregate Statistics

MetricValue
Total tasks39
Total human-equivalent hours1,167.5
Total Claude minutes789
Total supervisory minutes162
Total tokens4,546,500
Weighted average leverage factor88.8x
Weighted average supervisory leverage factor432.4x
Human-equivalent weeks29.2

Analysis

The highest factor of the day came in at 274.3x and the lowest at 15.0x, a spread of 18.3 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 160.0 of the 1,167.5 human-equivalent hours, or 14 percent of the day. No single task dominated the total, so the weighted average is representative.

Supervisory time was 162 minutes against 789 minutes of execution, a ratio of about 1 to 5. Supervisory leverage of 432.4x 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.