Resume / CV

Pre-built template tuned for parsing resumes and CVs into structured candidate profiles. Works on PDFs, Word docs, and image scans.

What gets extracted

Out of the box you get:

json

{
  "candidate": {
    "full_name": "Jane Doe",
    "email": "jane@example.com",
    "phone": "+1-555-0123",
    "location": "Brooklyn, NY"
  },
  "summary": "Senior backend engineer with 9 years of experience...",
  "skills": ["Go", "PostgreSQL", "Kubernetes", "AWS"],
  "experience": [
    {
      "company": "Acme Corp",
      "title": "Senior Engineer",
      "start_date": "2022-03",
      "end_date": null,
      "current": true,
      "highlights": [
        "Led migration from monolith to event-driven services.",
        "Reduced p99 latency by 60%."
      ]
    }
  ],
  "education": [
    {
      "institution": "University of Somewhere",
      "degree": "B.S. Computer Science",
      "graduated_year": 2016
    }
  ],
  "certifications": ["AWS Solutions Architect — Professional"],
  "links": {
    "linkedin": "https://linkedin.com/in/janedoe",
    "github":   "https://github.com/janedoe",
    "portfolio": null
  }
}

When to use

ATS pipelines and candidate sourcing.
Bulk-importing past candidate resumes into a CRM.
Anonymizing resumes (extract structured data → drop name + contact for blind review).

Customizing

Click Edit fields on the extraction page to add, remove, or rename fields. The template is just a starting point — your custom schema overrides the defaults.

Tips for accuracy

Use the original PDF if you have it. Resume PDFs have a text layer; images need OCR which is a hair less accurate.
One file = one candidate. Don't combine multiple resumes into one PDF.
The classifier helps at scale. Pair this template with a classification that routes only true resumes here — keeps cover letters and reference sheets out of your candidate stream.