Turn PDF resumes into structured candidate data

import csv
import pathlib
import time
import requests

BASE = "https://api.dataguru.cc/api/v1/documents"
HEADERS = {"Authorization": "Api-Key YOUR_API_KEY"}


def parse_resume(path):
    with open(path, "rb") as f:
        res = requests.post(f"{BASE}/resume-parser/jobs/", headers=HEADERS, files={"file": f}, timeout=120)
    res.raise_for_status()
    job_id = res.json()["job_id"]

    while True:
        job = requests.get(f"{BASE}/jobs/{job_id}/", headers=HEADERS, timeout=30).json()
        if job["status"] == "ready":
            return job["result"]
        if job["status"] == "failed":
            raise RuntimeError(job["error"])
        time.sleep(3)


with open("candidates.csv", "w", newline="") as out:
    writer = csv.writer(out)
    writer.writerow(["file", "name", "email", "phone", "skills", "strength", "completeness"])
    for pdf in sorted(pathlib.Path("resumes").glob("*.pdf")):
        r = parse_resume(pdf)
        writer.writerow([
            pdf.name, r.get("full_name"), r.get("email"), r.get("phone"),
            ", ".join(r.get("skills") or []), r.get("strength_score"), r.get("completeness_score"),
        ])

{
  "job_id": "3f6c1a2e-8b4d-4c1e-9a7f-2d5e6b7c8d9e",
  "parser_type": "resume-parser",
  "status": "ready",
  "result": {
    "full_name": "Jane Doe",
    "email": "[email protected]",
    "phone": "+1 555 0100",
    "skills": [
      "Python",
      "Django",
      "PostgreSQL",
      "AWS"
    ],
    "experience": [
      {
        "title": "Backend Engineer",
        "company": "Example Corp",
        "duration": "2021 - present"
      }
    ],
    "education": [
      {
        "degree": "BSc Computer Science",
        "institution": "Example University"
      }
    ],
    "strength_score": 82,
    "completeness_score": 90,
    "recommendations": [
      "Add measurable results to recent roles"
    ]
  },
  "error": null
}