POST /v1/jobs/search ACTIVE SERVICE

Job Discovery & Deduplication Service

Executes concurrent multi-platform discovery across LinkedIn and Naukri using direct internal RPC protocols. Automatically normalizes raw listings into the Common Job Schema, eliminates duplicate reposts via SHA-256 content hashing, and persists records into PostgreSQL.

1. Trigger Direct Multi-Platform Search

POST /v1/jobs/search

Initiates a live direct API search across connected LinkedIn and Naukri accounts.

Request Payload:

{
  "keywords": "Staff Software Engineer",
  "location": "Remote",
  "platforms": ["linkedin", "naukri"]
}

Response Payload:

{
  "ok": true,
  "count": 28,
  "jobs": [
    {
      "id": "job_172923982_x81b",
      "platform": "linkedin",
      "platformJobId": "3892819201",
      "title": "Staff Backend Engineer",
      "companyName": "Datadog",
      "location": "Remote",
      "applyType": "ONSITE_EASY_APPLY",
      "matchScore": 92,
      "matchExplanation": "Matched 5 skills (Node.js, PostgreSQL, Docker, Redis, REST APIs).",
      "dedupHash": "8f3b92...1c"
    }
  ]
}

2. List Discovered Jobs

GET /v1/jobs?platform=linkedin&applyType=ONSITE_EASY_APPLY

Query discovered listings with filtering by platform, apply type, keyword, or minimum match score.

Cross-Platform Deduplication Engine

Recruiters often post the same opening across both LinkedIn and Naukri with minor variations (e.g. "Google Inc." vs "Google LLC"). Our deduplication engine:

Normalizes company names by stripping corporate suffixes (Inc, LLC, Pvt Ltd, Technologies).
Generates a canonical SHA-256 hash combining normalized Title, Company, and Location.
Applies Dice-Sørensen word-set similarity scoring (>0.85 threshold) for duplicate detection across agencies.