An automated job discovery and cold outreach system built for software engineers. Finds fresh jobs daily, scores them against a configurable role profile, generates personalized LinkedIn outreach templates, and emails a curated list every morning.

Target user: Any developer running a job search — comes with presets for Backend Python (3+ YOE), Frontend React, and Fresher. Swap profiles to retarget the entire pipeline (search queries, scoring weights, outreach copy) without editing code.

Every day at 9:00 AM IST (automatic):

  1. Collects jobs from 4 sources + 150 company career pages
  2. Scores each job against the ACTIVE profile (title/tech/exp/signals)
  3. Drops anything below the profile's min-score, deduplicates,
     cleans up jobs not seen in 14 days
  4. For top 15 jobs:
       - Generates LinkedIn people-search URLs per company
         (titles configured on the profile — Eng Manager, Tech Lead,
         CTO, CEO, HR, etc.)
       - Writes a personalized DM using the profile's templates
  5. Emails the curated list to the recipient configured on the profile

Result: Candidate opens email, gets 15 fresh matching jobs with ready-to-send LinkedIn DMs. Takes ~20 min/day to work through. Change roles? Switch profile — no code changes.

The pipeline is driven by an active profile — a single config blob that controls search queries, scoring, location preferences, and outreach copy. Switching profiles re-targets the system end-to-end without touching code.

Stored as YAML in profiles/. Import any of them from the Profile page (or via POST /api/profiles/import):

A profile is one row in the profiles table (JSON config), with sections matching the Profile page tabs:

• Search — default search terms, positive/negative title keywords, relevant tech list, JSearch queries (country, posted-window, remote-only)
• Scoring — experience target (fresher / junior / mid / senior / any), min-score-to-show, min-score-to-store, weights for title/tech/experience/signals (sum to 100), core tech, domain signals
• Location — India-positive / India-negative keywords, timezone-compatible / incompatible lists
• Outreach — candidate name, bio, achievements, core/extra tech, short + long DM templates, LinkedIn search titles, email greeting, sender + recipient email, digest subject role word

• Sidebar lists all saved profiles and shows the currently-active one
• Import any YAML preset, edit it, then Save or Save & Activate
• Re-score All Jobs — recomputes every stored job's relevance against the active profile (optionally drops anything that now falls below min-score-to-store)
• Export YAML — download the active profile as YAML (round-trips back into profiles/)
• Duplicate / Delete — fork a profile to tweak, or remove an unused one (the active profile can't be deleted — activate another first)

Email digest fields living on the profile: recipient_email, email_greeting, email_digest_subject_role. recipient_email overrides RECIPIENT_EMAIL in .env, so different profiles can send to different inboxes.

The sender address is fixed to SENDER_EMAIL in .env and is not a profile setting — it has to match SENDER_APP_PASSWORD. The Profile page shows it read-only; change it in .env.

On first run, if no profile exists, the system seeds a Backend Python (legacy) profile from the pre-profile hardcoded settings and activates it — so existing installs keep working.

# 1. Install Python deps
pip install -r requirements.txt

# 2. Copy env template and fill in API keys
cp .env.example .env
# Edit .env with your API keys (see Environment Variables below)

# 3. Run the server
python -m uvicorn main:app --host 127.0.0.1 --port 8000

# 4. Open http://127.0.0.1:8000 in browser

Create a .env file in the project root with the following:

# ─── Required for JSearch (LinkedIn/Indeed/Glassdoor aggregator) ───
# Sign up: https://rapidapi.com/letscrape-6bRBa3QguO5/api/jsearch
# Free tier: 200 requests/month
RAPIDAPI_KEY=your_rapidapi_key_here

# ─── Required for Daily Email Digest ───
# Must use Gmail App Password, NOT your regular password
# Generate at: https://myaccount.google.com/apppasswords
# (Requires 2-Step Verification enabled first)
[email protected]
SENDER_APP_PASSWORD=abcdefghijklmnop
[email protected]

# ─── Daily Digest Timing (IST timezone) ───
DAILY_EMAIL_HOUR=9                  # 24-hour format (9 = 9:00 AM IST)
DAILY_JOBS_COUNT=15                 # Number of jobs per email

# ─── Optional: Hunter.io (not actively used — LinkedIn search replaces it) ───
# Sign up: https://hunter.io
HUNTER_API_KEY=

# ─── Optional: Google Sheets Export ───
# For n8n / external automation pipelines
# Setup: Create Google Cloud service account → download credentials.json
GOOGLE_SHEETS_CREDS=credentials.json
GOOGLE_SHEET_ID=

1. RapidAPI key (5 min)

   • Sign up at rapidapi.com
   • Subscribe to JSearch (Free Basic plan)
   • Copy X-RapidAPI-Key from dashboard

2. Gmail App Password (3 min)

   • Go to myaccount.google.com/security
   • Enable 2-Step Verification (required)
   • Go to myaccount.google.com/apppasswords
   • App name: Job Scraper → Create
   • Copy the 16-character password (ignore spaces)

3. Google Sheets (optional, 10 min)

   • See docs/02-setup.md for full walkthrough

See docs/02-setup.md for complete setup instructions.

• Backend: Python 3.13, FastAPI, SQLite, APScheduler
• Frontend: Vanilla JS, HTML, CSS (no framework) — three pages: Jobs, Outreach, Profile
• Config: YAML presets in profiles/ for role configurations; runtime config lives in the profiles SQLite table
• External APIs: JSearch (RapidAPI), Greenhouse, Lever, Ashby, Remotive, RemoteOK, Arbeitnow
• Email: Gmail SMTP with App Password

Everything free or near-free:

Total: $0–$5/month

Built for personal job search automation. Adapt freely for your own use.