American Reindustrialization — Discover Industrial Opportunities

Purpose

Discover companies, startups, suppliers, and job opportunities shaping the American reindustrialization wave on americanreindustrialization.com — a curated public directory spanning manufacturing, defense & security, energy, technology, materials & metals, aerospace, autonomous systems, industrial automation, logistics, supply chain, AI, and industrial software. Returns structured records: companies (logo, tagline, descriptions, HQ city/state, sector, tags, employee range, funding stage, founded year, careers email, website, jobs count) and jobs (title, description, employer, location, work mode, employment type, experience level, posting date, salary when published, and the external apply_url pointing to the employer's own ATS — Lever / Greenhouse / Ashby). Read-only; never submits, applies, or POSTs.

When to Use

• Bulk extraction of every company in a sector (e.g. "all U.S. fusion-energy companies", "all CNC machining suppliers", "all defense-tech startups in Texas").
• Job hunting / market mapping — "all senior roles in aerospace", "all on-site jobs in manufacturing posted in the last 30 days", "all open roles at Last Energy".
• Investor / corp-dev sourcing — enumerating early-stage reindustrialization companies by funding stage, sector, or geography.
• Building downstream tooling (newsletters, Slack feeds, dashboards) that watches the directory for new entries.
• Tag-based discovery (e.g. made-in-usa, fusion-energy, evtol, cnc-machining, sheet-metal-fabrication, predictive-maintenance).

Workflow

The site is a Vite/React SPA, but its content is served by a clean public JSON REST API at https://americanreindustrialization.com/api/* — no auth, no cookies, no anti-bot stealth, no rate-limit headers observed. The API is the canonical path; scripted browsing is strictly the fallback (the SPA shell is empty <div id="root"></div> until JS hydrates, so every browser-driven step pays a render-tax for data the API hands back in milliseconds). Cloudflare fronts the site but does not challenge the API. A residential proxy is not required.

1. Pick the right list endpoint

All list responses have the shape {"data": [...], "total": N, "page": 1, "limit": 20, "totalPages": K}. Detail endpoints return the bare object (no envelope). Enum endpoints (/api/categories, /api/tags, /api/jobs/categories, /api/jobs/companies, /api/jobs/tags, /api/jobs/titles) return a bare JSON array — no envelope, no pagination.

2. Filter /api/companies

Append as query string. Multiple filters combine with AND.

3. Filter /api/jobs

4. Walk pagination

GET /api/companies?status=published&limit=100&page=1   # → 68 results, totalPages=1
GET /api/jobs?limit=100&page=1                          # → 100/501
GET /api/jobs?limit=100&page=2                          # → next 100
…until page > totalPages

A Referer header is not required. No auth. Keep a courteous ≤ 2 req/s.

5. Read individual records

• GET /api/companies/{slug} returns the full company object — same field set as list items plus full categories[], tags[], company_categories[] (with is_primary), and products_services array.
• GET /api/jobs/{title-slug}-{8charId} returns the job with an embedded company object and the external apply_url (almost always an ATS URL like https://jobs.lever.co/..., https://boards.greenhouse.io/..., or https://jobs.ashbyhq.com/...).

The slug for a job is e.g. radar-intern-632a1249 — title-kebab + hyphen + an 8-char hex id. Get the slug from the list response's slug field, or from /companies/{slug} pages on the front-end (/jobs/l/{slug} href).

6. Resolve image URLs verbatim

logo_url / cover_image_url are pre-built absolute URLs on the img.americanreindustrialization.com CDN, the legacy americanreindustrialization.com/storage/... path, the company's own domain, or framerusercontent.com. Use them as-is.

Browser fallback

If the API ever returns 403/5xx (none observed during survey), drive the SPA directly:

1. browse open https://americanreindustrialization.com/{path} — wait ~2-3s for hydration; the bare HTML is empty until React mounts.
2. The following front-end URLs are deep-linkable and do honor URL params:
   • /companies?category={slug}&state={XX} — directory filter (note: front-end UI shows status=published data only).
   • /categories/{slug} — full company list for one category (single page, no pagination control).
   • /tags/{slug} — full company list for one tag.
   • /jobs?category={slug} — paginated job board, filtered by category.
   • /jobs?tag={slug} — paginated job board, filtered by tag.
   • /jobs/c/{category-slug} — single-page job list for one category (e.g. /jobs/c/aerospace → 99 listings; no pagination control).
   • /jobs/l/{slug} — single job detail; the green Apply on Company Site anchor's href is the same apply_url field.
   • /companies/{slug} — single company detail with sidebar View All Jobs at {Company} link.
3. browse get markdown body after each navigation; harvest data from the rendered text. The page-rendered text mirrors the API fields 1:1.
4. Do not attempt URL filters that the SPA reads only from in-page state — these are dead in the URL: ?company=<anything>, ?job_type=, ?work_mode=, ?location=, ?experience=, ?min_salary=, ?posted=, ?position=. They appear in the homepage's "X open roles" anchors (/jobs?company=anatar) but do not filter the list — they just navigate to the unfiltered job board. Use the API's company_id (UUID) instead.

Site-Specific Gotchas

• Public JSON API at /api/* is the canonical surface. Do not scrape the React-rendered HTML when the API can answer in one request. There is no anti-bot challenge on the API; no auth, no cookies, no Referer requirement. Cloudflare sits in front (Cf-Ray, Server: cloudflare) but does not block direct fetches.
• /api/companies total = 96 but /companies UI shows 68. The unfiltered API includes drafts / unpublished records. Add ?status=published (or ?published=true) to match the public catalog. This is the single most common mistake — leaving it off yields ~30% phantom records with stale or null fields.
• /api/jobs?company={name} and ?company_slug={slug} are silently ignored — they 200 with the unfiltered 501-job pool. Only company_id={uuid} filters. To find the UUID: GET /api/companies/{slug} → .id, or page through /api/jobs/companies (only 20 employers have active listings). Hardcoding company slugs in your filter logic will produce false-positive "all 501 jobs match this company" bugs.
• Filter param names are non-obvious. Companies use state= (not hq_state=) and query= (not q= or search=). Jobs use experience_level= (not experience= or level=) and job_type= in kebab-case (full-time; full_time returns zero). Unrecognized params are silently dropped — a 200 OK with total=501 always means "your filter didn't apply", never "no matches".
• work_mode=on-site is hyphenated. Enum values for work_mode: on-site (492), remote (6), hybrid (3). For experience_level: entry (18), mid (458), senior (21), lead (~4). For job_type: full-time (484), internship (15), part-time (1), contract (1).
• Front-end URL filters that DON'T work in the URL bar. The /jobs filter sidebar (Company, Position, Job Type, Work Mode, Experience, Location, Minimum Salary, Posted, Tag) is mostly UI-state only. Only ?category=, ?tag=, and the search-bar q= propagate to URL. The other dropdowns mutate React state and do not produce shareable / scriptable URLs — even though the homepage's "X open roles" badges link to /jobs?company=<slug> URLs, those don't actually filter the displayed list. Always use the JSON API for non-category/tag job filtering.
• /categories/{slug} and /jobs/c/{cat-slug} render full lists on a single page, no pagination. /companies and /jobs (the global views) do paginate (20 per page front-end, configurable up to ~100 on the API).
• jobs_count on company records is often 0 because most companies in the directory don't post jobs through the platform — only ~20 employers have active listings. Use /api/jobs/companies to enumerate hiring employers cheaply rather than counting company-by-company.
• apply_url points off-site. Every job's apply flow exits to the employer's own ATS (Lever, Greenhouse, Ashby, etc.). The site does not collect applications — never POST anywhere on americanreindustrialization.com for job applications.
• /api/news and /api/blog exist but currently return empty arrays ([]). Don't rely on them for content discovery yet.
• /api/states returns 404. No public enum endpoint for U.S. states; use a hardcoded list of two-letter abbreviations when needed, or harvest the distinct hq_state values across /api/companies.
• robots.txt is restrictive for AI crawlers — Cloudflare-managed Content-Signal: search=yes, ai-train=no and explicit Disallow: / for ClaudeBot, GPTBot, CCBot, Google-Extended, Bytespider, Applebot-Extended, Amazonbot, CloudflareBrowserRenderingCrawler. Honor it for compliant operation. The API does not check User-Agent; observed requests succeed regardless, but the site has expressly reserved AI-training rights under the EU Article 4 opt-out signaled in the robots header.
• Sitemap.xml is the cheap enumeration backstop. https://americanreindustrialization.com/sitemap.xml (~170 KB) lists every /companies/{slug}, /categories/{slug}, /tags/{slug}, and /jobs/c/{cat} URL — but does NOT list individual job-listing URLs (/jobs/l/...) or tag-scoped job views (/jobs/t/...). For job enumeration, walk /api/jobs pagination instead.
• No rate-limit headers were observed, but be courteous. Cloudflare is in front; aggressive bursts will earn 429s. ≤ 2 req/s sustained, batched paginated calls (limit=100) is the polite path.

Expected Output

Shape A — paginated company list (GET /api/companies?status=published&category=manufacturing&limit=100)

{
  "data": [
    {
      "id": "a1c19858-1eda-4bff-a0cb-d6f94ab9d3f7",
      "name": "Harmony AI (tryharmony.ai)",
      "slug": "harmony-ai",
      "tagline": "AI Automation for Manufacturing",
      "short_description": "Harmony AI provides an AI-native operating system that connects plant data...",
      "full_description": "Harmony AI is an AI-native operating system designed specifically for American manufacturing...",
      "website_url": "https://tryharmony.ai",
      "logo_url": "https://img.americanreindustrialization.com/company-assets/a1c19858-.../kt5XHThBvsz7....jpg",
      "hq_city": null,
      "hq_state": null,
      "hq_country": "United States",
      "founded_year": null,
      "employee_range": null,
      "funding_stage": null,
      "primary_sector": null,
      "is_us_based": true,
      "status": "published",
      "published_at": "2026-04-22T17:33:11.000000Z",
      "careers_email": null,
      "products_services": [],
      "jobs_count": 0,
      "categories": [
        { "id": "6e1ee891-...", "name": "Manufacturing", "slug": "manufacturing", "pivot": { "is_primary": 1 } }
      ],
      "tags": [
        { "id": "5923c8a5-...", "name": "AI & Machine Learning", "slug": "ai-machine-learning", "tag_type": "tech" }
      ]
    }
  ],
  "total": 36,
  "page": 1,
  "limit": 100,
  "totalPages": 1
}

Shape B — paginated job list (GET /api/jobs?category=aerospace&job_type=full-time&limit=100)

{
  "data": [
    {
      "id": "a1ba48c9-c18b-4bf4-9945-f82f1af481e5",
      "company_id": "a1ba45d1-3e7e-429c-8e13-3eb19eb4b3bc",
      "title": "Research Engineer, Aerosol & Liquid Dispersion Systems",
      "slug": "research-engineer-aerosol-liquid-dispersion-systems-9601c64b",
      "description": null,
      "job_type": "full-time",
      "work_mode": "on-site",
      "location_city": "El Segundo",
      "location_state": "CA",
      "location_country": "United States",
      "salary_min": null,
      "salary_max": null,
      "salary_currency": null,
      "salary_period": null,
      "experience_level": "mid",
      "apply_url": "https://jobs.lever.co/make-rain/...",
      "status": "published",
      "posted_at": "2026-05-08T04:31:23.000000Z",
      "expires_at": null,
      "is_auto_fetched": true,
      "company": {
        "id": "a1ba45d1-3e7e-429c-8e13-3eb19eb4b3bc",
        "name": "Rainmaker",
        "slug": "rainmaker",
        "logo_url": "https://img.americanreindustrialization.com/company-assets/.../dpyvw....jpg",
        "tagline": "Make rain. Make snow. Make..."
      }
    }
  ],
  "total": 73,
  "page": 1,
  "limit": 100,
  "totalPages": 1
}

Shape C — single company (GET /api/companies/relativity-space)

{
  "id": "...",
  "name": "Relativity Space",
  "slug": "relativity-space",
  "tagline": "...",
  "short_description": "...",
  "full_description": "...",
  "website_url": "https://www.relativityspace.com",
  "hq_city": "...",
  "hq_state": "...",
  "categories": [ { "name": "Aerospace", "slug": "aerospace", "pivot": { "is_primary": 1 } } ],
  "tags": [ { "name": "Space", "slug": "space" }, … ],
  "products_services": ["Terran R", "Stargate 3D printer", …],
  "jobs_count": 0,
  "careers_email": null
}

Shape D — single job (GET /api/jobs/radar-intern-632a1249)

{
  "id": "...",
  "company_id": "a1ba45d1-...",
  "title": "Radar Intern",
  "slug": "radar-intern-632a1249",
  "job_type": "internship",
  "work_mode": "on-site",
  "location_city": "Norman",
  "location_state": "OK",
  "experience_level": "entry",
  "apply_url": "https://jobs.lever.co/make-rain/58c47b24-5e07-4c40-84ca-046c040f5131",
  "posted_at": "2026-05-08T04:31:23.000000Z",
  "company": { "id": "...", "name": "Rainmaker", "slug": "rainmaker", "logo_url": "...", "tagline": "..." }
}

Shape E — taxonomy enum (GET /api/categories, GET /api/tags, GET /api/jobs/categories)

[
  {
    "id": "6e1ee891-4ade-4559-9ef4-56299c1c39aa",
    "name": "Manufacturing",
    "slug": "manufacturing",
    "description": "Companies involved in industrial manufacturing and production",
    "parent_id": null,
    "sort_order": 1,
    "image_url": null,
    "jobs_count": 253
  }
]

(Top-level array, no envelope, no pagination. jobs_count present only on /api/jobs/categories and /api/jobs/tags. tag_type present only on /api/tags and /api/jobs/tags — observed values include tech, industry, and others.)