Industry 4.0 job board connecting IT/OT engineering roles with hard-tech employers.
- docs/plan.md — product vision and what is shipped today.
- docs/supabase-environments.md — Supabase Cloud vs local Docker, and the production checklist.
- docs/future_plan.md — runbooks for upcoming work (Physical AI category, visa sponsorship, new ATS modules).
| Development (local) | Production | |
|---|---|---|
| Code runs on | Your machine | Vercel (frontend), GitHub Actions (scraper), Supabase Cloud (DB + Auth + API) |
| Database & API | npm run db:start → http://127.0.0.1:54321 |
https://<project-ref>.supabase.co |
| Frontend env | frontend/.env.local (gitignored) |
Vercel Environment Variables |
| Scraper env | scraper/.env (gitignored) |
GitHub Secrets + scheduled workflow |
| Auth emails | Mailpit at port 54324 | Real SMTP via Supabase Auth |
- Node.js 18+, npm
- Docker Desktop (or Docker Engine) running
- Python 3.11+ (only if you want to run the scraper)
cd frontend
npm install # installs the Next.js app and pins the Supabase CLI
npm run db:start # starts the full Supabase stack via DockerThe first run pulls images and applies supabase/migrations/*.sql + supabase/seed.sql. API ends up at http://127.0.0.1:54321.
npm run db:statusCopy the Publishable key (sb_publishable_…) from the output, then:
cp .env.example .env.local # in frontend/
# edit .env.local and paste the Publishable key after NEXT_PUBLIC_SUPABASE_ANON_KEY=NEXT_PUBLIC_SUPABASE_URL is already http://127.0.0.1:54321 in the template.
npm run dev # from frontend/Open http://localhost:3000. For magic-link login, read emails at http://127.0.0.1:54324 (Mailpit).
cp scraper/.env.example scraper/.env
# edit scraper/.env: paste the Secret key (sb_secret_…) and your OPENAI_API_KEY
cd scraper
uv venv && source .venv/bin/activate # Windows: .venv\Scripts\activate
uv pip install -r requirements.txt
playwright install chromium --with-deps
uv run main.py --test-run--test-run uses the first two companies and skips DB writes.
npm run db:reset # wipe + reapply migrations and seed
npm run db:stop # stop containers (data persists in Docker volume)The short version lives here; the full checklist (Supabase project setup, Vercel env vars, GitHub Actions secrets, custom domains) is in docs/supabase-environments.md.
- Supabase Cloud — create a project, apply
supabase/migrations/001_initial_schema.sql, copy the Project URL + anon key + service_role key. - Vercel — import the repo, set Root Directory to
frontend, addNEXT_PUBLIC_SUPABASE_URLandNEXT_PUBLIC_SUPABASE_ANON_KEY(anon only — never the service_role key here). Addhttps://<your-domain>/auth/callbackto Supabase Auth → Redirect URLs. - GitHub Actions — add
SUPABASE_URL,SUPABASE_SERVICE_KEY, andOPENAI_API_KEYas repository secrets. The workflow in.github/workflows/scraper.ymlruns daily at 06:00 UTC.
The service_role key (sb_secret_… locally, service_role in the Cloud dashboard) bypasses RLS. Keep it in scraper/.env and GitHub Secrets only — never in any NEXT_PUBLIC_* variable or frontend code.
| File | When | What goes in |
|---|---|---|
frontend/.env.local |
Local Next.js dev | NEXT_PUBLIC_SUPABASE_URL, NEXT_PUBLIC_SUPABASE_ANON_KEY (Publishable key) |
scraper/.env |
Local scraper run | SUPABASE_URL, SUPABASE_SERVICE_KEY (Secret key), OPENAI_API_KEY |
Both files are gitignored. Templates without secrets: frontend/.env.example, scraper/.env.example.
Production uses Vercel env vars and GitHub Secrets instead of committed files.
# Frontend
cd frontend && npm install && npm run dev
# Supabase local (pinned CLI, run from frontend/)
npm run db:start | db:stop | db:status | db:reset
# Scraper (local, full run)
cd scraper && source .venv/bin/activate && uv run main.pySee docs/plan.md for product context and docs/future_plan.md for upcoming work. Do not commit secrets.
Add a LICENSE file if you intend to open-source the project; until then, all rights reserved unless stated otherwise by the repository owner.