JobRader

Job market insights for job seekers — search live listings and see how demand and salaries move across skills.

JobRader pulls job listings from the Adzuna API, caches them locally, and records a daily snapshot of how many postings match each tracked skill. That history is what powers the market overview charts — Adzuna only exposes current listings, so the trend data has to be accumulated over time by our own scheduled job.

---

Tech stack

Layer	Choice
Frontend	Next.js 16, React 19, TypeScript, Tailwind CSS 4
Backend	Django 4.2
Database	SQLite, schema managed by hand via schema.sql
External data	Adzuna Jobs API

---

Features

• Market overview (/Market-Overview) — average salary by skill and a demand-over-time line chart, filterable by skill.
• Job listings (/dashboard) — free-text keyword and location search with paginated results, backed by a 30-minute cache so we don't burn through the Adzuna rate limit.
• Job detail (/jobs/[id]) — full description for a single posting, read from the local cache.
• Me (/me) — favourited jobs and saved searches.
• Login / Register (/login) — UI is in place; not yet wired to a backend auth endpoint.

---

Project structure

.
├── backend/                     # Django project
│   ├── config/                  # settings, urls, wsgi/asgi
│   ├── jobs/                    # the only app
│   │   ├── models.py            # JobPosting, DemandSnapshot (managed = False)
│   │   ├── services.py          # Adzuna client + normalisation + upsert
│   │   ├── views.py             # job_search, job_detail, market_overview
│   │   └── management/commands/
│   │       └── fetch_jobs.py    # scheduled batch fetch + daily snapshot
│   ├── schema.sql               # source of truth for the DB schema
│   └── requirements.txt
└── frontend/                    # Next.js app
    └── app/
        ├── api/jobs/            # route handlers proxying the Django API
        ├── components/Sidebar.tsx
        ├── Market-Overview/
        ├── dashboard/
        ├── jobs/[id]/
        ├── login/
        └── me/

---

Getting started

Prerequisites

• Python 3.11+
• Node.js 20+
• An Adzuna API account (free) for an app_id and app_key

1. Backend

cd backend
python -m venv venv
source venv/bin/activate        # Windows: venv\Scripts\activate
pip install -r requirements.txt
pip install django-cors-headers # not yet pinned in requirements.txt, but settings.py needs it

Create backend/.env:

ADZUNA_APP_ID=your_app_id
ADZUNA_APP_KEY=your_app_key
ADZUNA_COUNTRY=au

Create the database. The jobs models are declared managed = False, so migrate will not create job_postings or demand_snapshots — schema.sql does:

sqlite3 db.sqlite3 < schema.sql   # app tables
python manage.py migrate          # Django's own tables (auth, sessions, admin)
python manage.py runserver        # http://127.0.0.1:8000

2. Frontend

cd frontend
npm install
npm run dev                       # http://localhost:3000

The Next.js route handlers proxy to Django at http://127.0.0.1:8000/api/jobs/ by default. Override with DJANGO_API_URL in frontend/.env.local if your backend runs elsewhere.

3. Seed the trend data

The market overview needs at least one snapshot to render anything:

cd backend
python manage.py fetch_jobs
python manage.py fetch_jobs --location "Perth" --date 2026-07-01

Flag	Default	Notes
--location	Western Australia	Passed to Adzuna for every tracked skill
--date	today	Labels the snapshot; Adzuna only returns current listings, so this backdates the label, not the data

Tracked skills live in TRACKED_SKILLS at the top of fetch_jobs.py. In production this should run once a day on cron or Celery beat — never on a page load.

---

Environment variables

Variable	Where	Required	Default
ADZUNA_APP_ID	backend/.env	Yes	''
ADZUNA_APP_KEY	backend/.env	Yes	''
ADZUNA_COUNTRY	backend/.env	No	au
DJANGO_API_URL	frontend/.env.local	No	http://127.0.0.1:8000/api/jobs/

---

API reference

All endpoints are GET only and return JSON.

GET /api/jobs/

Search jobs. Serves from an in-memory cache first, then Adzuna, then falls back to whatever rows are already in job_postings if Adzuna is unreachable.

Query param	Default	Description
q	''	Keyword
location	''	Location
page	1	20 results per page

{
  "jobs": [
    {
      "id": "5312345678",
      "source": "adzuna",
      "title": "Junior Software Engineer",
      "company": "Example Pty Ltd",
      "location": "Perth, WA",
      "salary_min": 75000,
      "salary_max": 90000,
      "url": "https://www.adzuna.com.au/...",
      "posted_at": "2026-07-28T09:14:00Z",
      "category": "IT Jobs"
    }
  ],
  "page": 1,
  "total_pages": 12,
  "total_count": 231,
  "cached": false
}

GET /api/jobs/<external_id>/

Full detail for one posting, including description. Reads only from the local cache, so it 404s for an id that was never returned by a search.

GET /api/market-overview/

Aggregates for the charts.

{
  "salary_by_skill": [{ "skill": "IT Jobs", "salary": 105000 }],
  "demand_by_skill": {
    "Python": [{ "date": "2026-07-29", "count": 412 }]
  }
}

salary_by_skill averages the midpoint of salary_min/salary_max grouped by Adzuna category. demand_by_skill is the accumulated demand_snapshots history, grouped by skill and ordered by date.

---

Data model

Defined in schema.sql. users, favourite_jobs, and saved_searches exist in the schema but aren't wired to Django models yet.

Table	Purpose
job_postings	Cached listings, upserted on (source, external_id)
demand_snapshots	One row per skill per day: count of matching postings
users	Email + password hash
favourite_jobs	Saved jobs per user
saved_searches	Saved keyword/location filters, with an alert flag

---

Team members

Name	GitHub	Responsible for
Soyeon Kim	@soyeonkim00	Sidebar navigation, login / register page
Lee Jon	@LeeJon94	Me page — favourites and saved searches
Yupei	@zyp-gamer	Market overview page — salary and demand charts
Angie Lee	@dldmewl2970	Job search page — search, filters, pagination