SPB Git forge
spboucher.ai ↗

spb/airiskindex

Public

The most methodologically rigorous, fully transparent AI job-exposure index.

www.airiskindex.io
webappaiindexlaborresearch
21commits 1branches 0releases
251.0 KBsize
maindefault branch
20 days agolast push
TypeScript 88% Python 6.1% SQL 2.7% CSS 1.2% JavaScript 0.9% Shell 0.8%

page 1

  1. Fri, Sep 4, 2026 1

    1. chore(retrait cluster): synchro copie du nœud M3U96b + README « URL live hors fonction » (2026-09-04)
      Simon-Pierre Boucher committed 20 days ago · 1 file +2

  2. Wed, Aug 5, 2026 20

    1. feat(web): insights dashboard — scatter (substitution × augmentation), distribution, wage deciles, group bars, top-20 table
      …
      - ScatterPlot: 923 occupations, dot area ~ employment, quadrant
  annotations, native tooltips, hairline grid (server-rendered SVG)
- DistributionChart reuse with employment-weighted-mean marker
- DecileColumns: employment-weighted substitution per wage decile
- employment-weighted SOC group bars + top-20 table with CI/wage/employment

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 2 files +315 −3
    2. feat(web): distribution charts on detail pages + ultra-detailed README with metric badges
      …
      - DistributionChart: task histogram + position-in-index (emphasis form,
  recessive context bins, accent marker with rank pill)
- README: badges, headline metrics table, methodology summary, API docs,
  monorepo map, quick start, roadmap, citation & contact

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 3 files +290 −30
    3. feat(web): sharper detail page — sub-score dot plot on shared axis, weight pills; fix insights int overflow
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 3 files +117 −31
    4. feat(web): full-index page (paginated by 50) + insights page with headline metrics
      …
      - /ranking: all scored occupations ranked, previous/next pagination,
  CI + sub-scores + wage/employment per row
- /insights: hero figure (share of US wage bill under substitution
  pressure), employment-weighted tiles, augmentation counterpoint,
  most-workers-at-high-pressure and widest-CI lists, reading guide
- nav/footer links updated

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 4 files +402 −3
    5. fix(worker): recompute loads per-occupation with minimal selects (avoid 1GB payload overflow)
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 1 file +25 −12
    6. feat(web): SVG brand logo (ranked score bars) + favicon via app/icon.svg
      …
      LogoMark wired into header, footer and mobile drawer; full logo with
CI-whisker signature in public/logo.svg.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 5 files +53 −12
    7. chore: add file headers (author, contact, copyright, description) to all source files
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 50 files +457
    8. fix(web): mobile layout — Tailwind content missed components/, truncate min-w-0 overflow, drawer portal
      …
      - tailwind.config content now scans components/ and lib/ (classes unique to
  those files were never generated: drawer fixed/inset/animation, search
  dropdown styling)
- min-w-0 on truncate flex children + ranking grid children: page no longer
  overflows horizontally, mobile browsers no longer zoom out
- drawer rendered via portal to <body> (blurred sticky header is a containing
  block for fixed descendants)

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 5 files +27 −17
    9. feat(web): mobile menu as right-side drawer with slide-in animation
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 2 files +34 −3
    10. feat(web): mobile dropdown menu (hamburger) in sticky header
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 2 files +80 −2
    11. feat(web): design overhaul — sharper modern light design, fully mobile-responsive
      …
      Sticky blurred header, hero wash, pill kicker, card idiom with hover
elevation, refined footer (3-column), responsive paddings/typography,
scrollable audit tables on small screens, explicit viewport/themeColor.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 10 files +195 −78
    12. feat(web): full methodology documentation page + terms of service + privacy policy
      …
      Operator: Simon-Pierre Boucher, contact@spboucher.ai. Methodology page pulls
weights/version live from the scoring engine; footer links added.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 4 files +698 −2
    13. fix(worker): tolerate truncated/invalid batch results; raise rater max_tokens
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 1 file +15 −2
    14. feat(worker): parallel per-model rating batches with resume via state file
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 1 file +32 −8
    15. infra(deploy): track PM2 wrapper scripts in repo so rsync --delete cannot remove them
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 2 files +7
    16. feat: full O*NET universe — real ETL, deep detail pages, search, browse, rating scripts
      …
      - etl: OEWS wage download, O*NET 30.3 transform (occupations/tasks/importance),
  wages transform, psycopg loader; 1,016 occupations + 18,796 tasks loaded
- web: live search, /occupations browse grouped by SOC major group, home with
  hero stats + top/least exposed + group averages; detail page adds wages,
  employment, rank/percentile, weighted dimension breakdown, full per-task
  rater audit trail (model/rating/rationale), related occupations
- worker: one-shot rate script (batch submit/poll/ingest, --limit pilot mode),
  purge-demo script, thinking disabled for the rater
- layout: preview banner now driven by whether the latest run is demo-based

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 16 files +1,088 −196
    17. docs: update deploy section to actual m3u96b layout (PM2 + brew services, no systemd/Docker)
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 1 file +20 −14
    18. feat(web): ranking home page + occupation detail pages with demo dataset
      …
      - ranking: substitution leaderboard with CI whiskers, hover tooltips,
  sr-only table fallback; sequential single-hue marks per dataviz specs
- detail: sub-score stat tiles with interval strips, highly-exposed-tasks
  meter, per-task bars with CI, visible data-table fallback,
  adaptation-not-doom framing copy
- seed: 10 real O*NET occupations x 3 tasks with deterministic synthetic
  two-model panel ratings (demo-panel-a/b); preview banner in layout
- light + dark viz roles from the validated reference palette

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 6 files +756 −60
    19. feat(db): initial migration
      …
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 2 files +172
    20. feat: scaffold AI Risk Index platform (research corpus, methodology v1 draft, monorepo)
      …
      - docs/research: 4-document research corpus grounding v1 (indices, data
  sources, rater API, landscape/evidence), compiled 2026-08-05
- docs/methodology: METHODOLOGY.md v1 draft, changelog, worked example
- packages/scoring: pure deterministic engine (5 weighted dimensions,
  barriers inverted, exposure/substitution/augmentation sub-scores with
  CI bounds), snapshot + fast-check property tests
- packages/db: Prisma schema with rating audit trail and immutable runs
- apps/web: Next.js 14 site + public API v1 (health, methodology,
  occupations)
- apps/worker: BullMQ multi-model rater pipeline (Message Batches,
  versioned rubric prompt v1) + score:recompute
- apps/etl: Python skeleton with O*NET 30.3 downloader
- infra: compose files, deploy.sh, ngrok.yml, env template

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher committed 1 mo ago · 74 files +7,101