Skip to content

Resume Screening and Shortlisting

Example prompt: "When resumes are uploaded to the 'Applications' Google Drive folder, score each candidate against our job description, rank the top 10, and put the shortlist in a Google Sheet."

How to automate resume screening with GloriaMundo

The Problem

Hiring for a single role can generate dozens or even hundreds of applications. A recruiter has to open each resume, read through it, mentally compare qualifications against the job description, and decide whether the candidate is worth progressing. This initial screen is repetitive, time-consuming, and prone to inconsistency — the 50th resume reviewed on a Friday afternoon does not get the same attention as the 5th one on Monday morning. For small HR teams handling multiple open roles, the screening backlog can delay the entire hiring timeline by days.

How GloriaMundo Solves It

We build a workflow that triggers when new files appear in a designated Google Drive folder. An integration step retrieves each uploaded resume. A code step extracts the text content from PDFs and Word documents so the AI can read them. An LLM step reads each resume against the job description you have provided and scores the candidate on key criteria — relevant experience, required skills, education, and overall fit — producing a numerical score and a short justification for each. A second code step ranks all scored candidates and selects the top 10. Finally, an integration step writes the ranked shortlist to a Google Sheet with scores, summaries, and links to the original files. Glass Box preview shows you the scoring criteria and sample evaluations before the full batch runs, so you can adjust the weighting if needed.

Example Workflow Steps

  1. Trigger (integration): Fires when new files are uploaded to the 'Applications' folder in Google Drive.
  2. Step 1 (integration): Retrieve the uploaded resume files from Google Drive.
  3. Step 2 (code): Extract text content from PDF and Word documents into plain text for analysis.
  4. Step 3 (LLM): Score each candidate against the job description on criteria including relevant experience, required skills, qualifications, and overall fit. Produce a numerical score and a two-sentence justification.
  5. Step 4 (code): Rank candidates by score and select the top 10 for the shortlist.
  6. Step 5 (integration): Write the ranked shortlist to a Google Sheet — candidate name, score, summary, and a link to the original resume.
  7. Step 6 (integration): Send a Slack notification to the hiring manager that the shortlist is ready for review.

Integrations Used

  • Google Drive — source of uploaded resume files
  • Google Sheets — destination for the ranked shortlist with scores and summaries
  • Slack — notifies the hiring manager when the shortlist is ready

Who This Is For

Recruiters and HR managers who handle high-volume hiring and need a consistent, fast way to screen applications without sacrificing thoroughness. Particularly useful for roles that attract 50 or more applicants. Since this workflow processes personal data from candidates, use the Glass Box preview to verify what is being stored and shared before running.

Time & Cost Saved

Manually screening a single resume takes roughly 3-5 minutes. For a role with 80 applicants, that is 4-7 hours of screening work. This workflow reduces the effort to reviewing a pre-ranked shortlist of 10 candidates, which takes around 20-30 minutes. Over a hiring cycle with multiple open roles, a small HR team could save 15-25 hours. The workflow uses integration, code, and LLM steps, costing a few credits per batch of resumes.