Thirty-eight tasks. May 9, 2026 weighted to 26.9x leverage across 632.5 human-equivalent hours in 1,410 Claude-minutes. The day was a pre-launch sweep across iOS web parity, an end-to-end status site stand-up, a fleet-wide accessibility audit fix, an analytics platform overhaul, and a marketing-site canon-swap propagation. Supervisory leverage closed at 223.2x.
The volume reflects a launch deadline; 15.8 weeks of human-equivalent throughput in twenty-three and a half hours of Claude wall-clock. The 85.7x ceiling came from an 8-phase mobile rebuild rebuilding the mobile client to match the web client, while the floor in the table sits at 6.7x on a four-tab settings restructure with extensive design-token migration. The middle of the distribution is dominated by accessibility audits, content-pipeline integrity work, and the ground infrastructure for the launch site.
About These Records
These time records capture personal project work done with Claude Code (Anthropic) only. They do not include work done with ChatGPT (OpenAI), Gemini (Google), Grok (xAI), or other models, all of which I use extensively. Client work is also excluded, despite being primarily Claude Code. The actual total AI-assisted output for any given day is substantially higher than what appears here.
Task Log
| # | Task | Human Est. | Claude | Sup. | Factor | Sup. Factor |
|---|---|---|---|---|---|---|
| 1 | iOS web-parity rebuild: 8 phases ; phase machine restructure, an app shell+a top-nav component, launch routing fix, HomeView (slim hub), multi-course Dashboard, CoursesView+CourseDetailView, SettingsView split, container/transitions/radius polish | 50.0h | 35m | 8m | 85.7x | 375.0x |
| 2 | an analytics platform: date-range fix + SSE-driven realtime ticks + bounce/duration/GeoIP + funnel ordering, attribution models, webhook handlers, CSV export, IP exclusions, public dashboard share | 60.0h | 44m | 6m | 81.8x | 600.0x |
| 3 | a status site: built and deployed a status site end-to-end ; new clients/a status site SPA (React 19/Vite/TS), a monitoring tool schema + public read API + alembic migration + 12 sanitization tests, admin-service banner.channels JSONB + public banners endpoint, | 80.0h | 70m | 8m | 68.6x | 600.0x |
| 4 | iOS Help fixes: tab strikethrough fix (overlay alignment), port 40 help guides verbatim from a help-doc source file → a help-doc target file (sidebar+content layout, iPhone sheet), embed 5 legal docs (privacy, terms, accessibility, trademarks, | 24.0h | 22m | 4m | 65.5x | 360.0x |
| 5 | iOS app facelift: SwiftUI design system port (tokens, typography, 14 components), a brand sans font bundling, a design theme shim, migrate 6 high-traffic views (LoginView, ResultsView, WelcomeView, ProfileView, DashboardView, BugReportView), pbxproj patch, docs | 40.0h | 50m | 6m | 48.0x | 400.0x |
| 6 | iOS Settings/Profile/Help web parity: fix sign-in button (a top-nav component overflow on iPhone), build new HelpView (5-tab Overview/FAQ/Guides/WhatsNew/Legal + .help phase + bug-report bridge), refactor ProfileView into hero+4-tab (Profile/Resume/Subscription/Account), | 18.0h | 24m | 5m | 45.0x | 216.0x |
| 7 | Build reusable static-site Terraform module (S3+OAC+CloudFront+ACM+Route53) with edge-enforced CloudFront-Function an access gate gate, plus english-accelastudy-website root stack (prod imports existing E51I2L5WDXNNS via auto-discovering import.sh, staging fresh-provisions with the gate). | 9.0h | 14m | 4m | 38.6x | 135.0x |
| 8 | Upgrade a language-exam product site (a language-proficiency exam product) to multi-page subscription product site: standalone /pricing/ page with comparison table & FAQ, switched nav to standalone routes, live header CTA, sister-site parity in Custom block, README + CHANGELOG updated. Verified clean build (26 pages, | 5.0h | 8m | 3m | 37.5x | 100.0x |
| 9 | Major engine fix + audit expansion. (1) Built a backfill script - deterministic pairid linkage backfill across 234 synthesized domain packages. Drove pairid coverage from 32.3% to 54.1% across 1.29M questions, with the worst cert domains (a professional cert 0.1%->38.1%, a professional cert, a professional cert, | 16.0h | 28m | 6m | 34.3x | 160.0x |
| 10 | a status site: round 2 ; closed remaining gaps from initial deploy. a monitoring tool frontend SiteSettingsForm gets publicstatusvisible/group/publicdisplayname/publicdescription fields; new IncidentDetailModal lets operators set severity/title/publicvisible and post markdown updates (investigating→identified→mon... | 32.0h | 65m | 2m | 29.5x | 960.0x |
| 11 | MEDIUM cleanup wave: 9 reduced-motion guards + 8 sr-only utilities + 594 h1->h2 codemod demotions across 241 files + 50 input-adjacent-label codemod pairings + 5 hand-fixes (BillingPage h1, Blog.jsx h1s, purchase-service globals.css, a legacy product site SCSS sr-only, charlessieg-redesign exemption); | 14.0h | 30m | 1m | 28.0x | 840.0x |
| 12 | Fleet-wide a11y fix sweep across 56 UI repos: 2,460 HIGH findings fixed (2,235 via a simulator suite label-pairing codemod + 17 manual + 5 wave-1 activities-react + 59 wave-3 client apps + 136 wave-4 tools fleet + 8 a simulator suite primitives); | 60.0h | 130m | 5m | 27.7x | 720.0x |
| 13 | Final wave: clear remaining 192 HIGH a11y findings ; patched 71 stale cloudops dist HTML files with lang=en (Python sed), dispatched focused subagent to fix 118 of 120 a simulator suite view-level inputs/svgs/clickable-divs (NetworkTopology+PolicyEditor+PacketInspector+ProjectBoard+a top-nav component+30 more dashboard... | 16.0h | 35m | 1m | 27.4x | 960.0x |
| 14 | Port web Help Center guide articles to iOS (40 docs, 7 categories) and rebuild Guides tab with sidebar+content layout | 8.0h | 18m | 5m | 26.7x | 96.0x |
| 15 | Drafted Making a learning platform Accessible to All across 3 sites ; a personal site (~3500-word technical deep-dive with mermaid wave diagram + 6 reference tables + concrete codebase counts: 2185 TSX/JSX files, 2527 native buttons, 3486 form inputs, 3019 ARIA uses, 1530 aria-labels, 752 aria-hidden, 101 role=button, | 12.0h | 28m | 2m | 25.7x | 360.0x |
| 16 | Drove the a structured-content spec catalog spec audit from 257 LOW (post-prior-pass) to absolute zero across all four severities. Tightened a spec auditor (broadened verb whitelist, fixed cross-domain prefix detection, normalized weight-sum auto-fix to handle any non-100 sum, relaxed cross-domain check to >=1, | 14.0h | 35m | 4m | 24.0x | 210.0x |
| 17 | Refactor CLAUDE.md chain: extract patent checklist, repo map, ADR rules, SSM, domain inventory, synthesis pipeline into subtree files; relocate API keys to mode-600 env file outside prompt | 2.5h | 7m | 3m | 21.4x | 50.0x |
| 18 | Built deterministic Python accessibility-audit checker (15 rules, 56-repo discovery, brace/quote-aware JSX tokeniser, JSON+MD output, mode-aware exit codes); updated accessibility-audit.md with Phase 0 spec citing the script; ran fleet-wide audit (414 HIGH + 2553 MEDIUM identified); | 24.0h | 70m | 4m | 20.6x | 360.0x |
| 19 | Content audit reconciliation: dropped 30 of 40 findings (all 11 CRITICALs + all 14 catalog/canonical MEDIUMs + 5 LOWs). Wrote a dedup script to remap 33 collided exam_code values to vendor-correct codes (a professional cert Plus suite -> a professional cert/a professional cert/a professional cert/a professional cert/a ... | 4.0h | 12m | 2m | 20.0x | 120.0x |
| 20 | Restructure iOS ProfileView to mirror web Profile 4-tab layout (hero + Profile/Resume/Subscription/Account) | 6.0h | 18m | 5m | 20.0x | 72.0x |
| 21 | an analytics platform last-hour delta on MetricCards (backend + frontend), an admin tool SoundProvider + priority-aware notification/anomaly cues, and tool-specific cues across foundry/chronicle/trellis/herald/meridian/envoy/fulcrum/tribe (plus pre-existing tsc fixes) | 18.0h | 55m | 5m | 19.6x | 216.0x |
| 22 | Sound effects follow-ups: nuked an inference engine root node_modules + made tools self-contained, Terraform stack lib-pipelines/ provisioning a build service + a CI/CD pipeline for all 6 publishable @avian/* libs (5 imported + new sound-effects), tool-specific cues wired into chirp/courier/vigil/slate/packed | 16.0h | 50m | 4m | 19.2x | 240.0x |
| 23 | Build CoursesView and CourseDetailView for a learning platform iOS app (web parity) | 6.0h | 22m | 8m | 16.4x | 45.0x |
| 24 | Port web legal docs into iOS HelpView ; embed privacy policy, terms, accessibility, trademarks, and credits as scrollable in-app a legal-doc tree node tree; rebuild Legal tab with iPad sidebar and iPhone sheet flow | 6.0h | 22m | 5m | 16.4x | 72.0x |
| 25 | Sound effects fleet rollout: created a shared sound-effects library v0.1.0 standalone package, uploaded 28 mp3s to an assets CDN CDN with CORS, wired SoundProvider into 21 tools (3 needed manual handling, fixed pre-existing TS/JSX errors in cadence/courier/dossier along the way), committed + pushed each tool repo. | 24.0h | 90m | 8m | 16.0x | 180.0x |
| 26 | an analytics platform: webhook hardening (require pulsesiteid, no default fallback) + dedicated 30 req/min webhook rate limit; deployed and verified live | 4.0h | 16m | 2m | 15.0x | 120.0x |
| 27 | Mode-1 a11y audit on a web client + 6 React libs: 11 HIGH fixes (RemoteBanners aria-live, HelpCenter dialog focus, LabConsole tab pattern, ProgressBar/ExamScoreReport/Sidebar progress+log roles, ProceduralStepSequencing focus-visible, InteractiveMap :focus-visible, BugReportModal dialog semantics, | 12.0h | 55m | 3m | 13.1x | 240.0x |
| 28 | a marketing site staging: redeploy with an access gate, restore real home page + /platforms/ nav link, remove /vote teaser; ship stage-isolated dist/ | 7.5h | 36m | 4m | 12.5x | 112.5x |
| 29 | Build real SettingsView for iOS app mirroring web settings page (appearance, language, study prefs, voice, accessibility, privacy, about sections) | 3.0h | 15m | 5m | 12.0x | 36.0x |
| 30 | Migrate a SwiftUI view to an inference engine iOS design system (design tokens, design typography, a button component, a card component, an empty-state component) | 1.5h | 8m | 3m | 11.2x | 30.0x |
| 31 | Rebuild a SwiftUI view to multi-course portfolio matching web Dashboard.tsx | 4.0h | 22m | 5m | 10.9x | 48.0x |
| 32 | Migrate a SwiftUI view (1521 lines) to an inference engine iOS design system tokens, typography, and components | 3.0h | 18m | 3m | 10.0x | 60.0x |
| 33 | a marketing site canon-swap propagation: replace hardcoded counts with [[canon:...]] placeholders across press, about, how-it-works, faq, accessibility, pricing, courses, free, 5 feature pages and shared pricing-card partial; fix stale patent counts (27→29 filings, 593/613→637 claims); | 5.0h | 35m | 2m | 8.6x | 150.0x |
| 34 | canon-swap sweep across 7 sister sites: mcat/lsat/ap/test-prep/english (OtherProducts blocks), a corporate site corporate (10 files - patent counts on index/about/ip/timeline/products/dossier/etc), enterprise (activity-formats); fix stale 27/28 filings → 29 + 593/613 claims → 637 + 20→13 activity formats; | 4.0h | 30m | 1m | 8.0x | 240.0x |
| 35 | Activities catalog reorg (default+5 addons across 62 categories) + 4 web bug fixes (data-driven Service Match applicability, Privacy footer link, Bio Profile→Resume, unenroll→autopilot cascade) + Settings prefs gray-out ; three repos committed and pushed | 14.0h | 110m | 18m | 7.6x | 46.7x |
| 36 | pre-launch staging audit + fixes: add og:image fallback in _metadata.jinja, strip /index.html from canonical URLs, add Exclude/ExcludeWhere collection filters in narrative CMS (4 unit tests), exclude a deferred category + a deferred category categories + 85 child course pages from rendering, | 5.0h | 40m | 2m | 7.5x | 150.0x |
| 37 | Migrate a SwiftUI view (1108 lines) to an inference engine iOS design system ; replace a design theme tokens with design tokens, update typography to design typography presets, replace ad-hoc cards/buttons with a card component/a button component/a badge component/an empty-state component/an inline-alert component, | 3.0h | 25m | 3m | 7.2x | 60.0x |
| 38 | Restructure a SwiftUI view to 4-tab layout (General/Autopilot/Accessibility/Privacy) with Audio section, extended-time toggle, Privacy Policy link, and SoundManager integration | 2.0h | 18m | 5m | 6.7x | 24.0x |
Aggregate Statistics
| Metric | Value |
|---|---|
| Total tasks | 38 |
| Total human-equivalent hours | 632.5 |
| Total Claude minutes | 1410 |
| Total supervisory minutes | 170 |
| Total tokens | 6,445,000 |
| Weighted average leverage factor | 26.9x |
| Weighted average supervisory leverage factor | 223.2x |
| Human-equivalent weeks | 15.8 |
Analysis
The day's leverage distribution matters more than the headline figure. The 85.7x ceiling came from iOS web-parity rebuild: 8 phases ; phase machine restructure, an app shell+a top-nav component,; the 6.7x floor was Restructure a SwiftUI view to 4-tab layout (General/Autopilot/Accessibility/Privacy) with Audio sect.... Tasks at the top of the distribution share a shape: tightly-scoped specifications, clear success criteria, and minimal integration ambiguity. The AI doesn't need to discover anything new; it executes against an explicit target.
Tasks at the bottom run differently. They're either bounded by review-heavy work where every step gets verified, or they involve ambiguity that demands several rounds of trial and adjustment. The factor is real and informative, not a failure mode.
The supervisory leverage figure (223.2x today) tracks something orthogonal to wall-clock leverage. It's the ratio of human-equivalent output to human prompt-writing time. It stays high even on lower-leverage days because supervisory minutes scale with task count, not with the human-hour estimate; a 20-minute task and a 4-hour task can both be specified in two minutes of human prompt-writing.
Across the 38 tasks, the day produced roughly 15.8 weeks of senior-engineer-equivalent throughput in 23.5 hours of model wall-clock. That ratio is the practical answer to the question of how much output a single operator can move per day when the model handles the execution and the operator handles the direction.