Score and rank job applicants from CV uploads with Gemini

Find Your AI Training Job
Chat with DigiNo Toucan and get matched to remote roles in one minute.

Recruitment agencies and HR teams waste hours reading CVs that never had a chance. This AI automation handles the first pass for them, scoring every applicant against a defined job description and logging the results instantly.

What This Automation Does

Accepts CV submissions from candidates through a public web form that captures name, email, and a PDF upload
Extracts readable text from each PDF using Mistral's OCR model so the AI receives clean, accurate content
Sends the extracted CV text to Google Gemini with a structured hiring prompt, returning a qualification score and written explanation
Writes every applicant's name, email, score, and AI explanation into a Google Sheets dashboard as each application arrives

Tools Used

n8n
Mistral
Google Gemini
Google Sheets

Work faster

Stop Typing and Start Speaking With Typeless

Typeless turns what you say into clean, formatted text in any app on Mac, Windows, Linux, iOS and Android, which it rates at 4x faster than typing. It strips filler words, keeps only your final phrasing when you change your mind mid-sentence, and works in 100+ languages. Emails, AI prompts, client messages and docs, done by voice.

Try Typeless →

Where to Get Hired for This Skill

On Contra, top freelancers across this stack have earned 130 combined verified reviews from real client projects.

Source: Contra freelancer search · refreshed 30 May 2026

Start Earning as a Freelancer on Contra

Contra is a commission-free professional network for independents. Browse live AI automation work and keep what you earn.

Join Contra Free →

How To Build It

Build the candidate intake form

A public-facing web form collects each applicant's name, email address, and CV as a PDF upload, giving your client a shareable link they can drop straight into a job posting.

Log every applicant on arrival

The moment a form is submitted, the candidate's name and email are written to the Google Sheets dashboard before any AI processing begins, ensuring no applicant is lost even if a downstream step fails.

Extract clean text from each PDF

Mistral's OCR model processes the uploaded PDF and returns accurate, structured text from the CV, handling scanned documents and formatted layouts that plain text parsers typically mangle.

Score the CV against the job description

The extracted CV text is sent to Google Gemini with a prompt that instructs it to act as a hiring assistant, producing a numerical qualification score and a written rationale based on the specific role requirements your client defines.

Write structured results to the dashboard

A JSON output parser captures the AI's score and explanation in a reliable, consistent format, then pushes both fields into the corresponding row in Google Sheets so the client sees a complete, sortable applicant record.

Pitfalls

Gemini's prompt is highly sensitive to how the job description is written. Vague or generic role requirements produce vague scores. Build in a prompt review step with your client during onboarding so the scoring criteria are specific before any real CVs flow through.
Google Sheets has a per-minute write quota that trips when a client shares the form link publicly and receives a surge of applications. Batch writes or a short delay between rows prevents quota errors from silently dropping applicant records.
Mistral OCR handles most PDFs well but struggles with image-heavy CVs where the text is embedded in graphics or tables rendered as images. Warn clients that candidates submitting design-portfolio-style CVs may return incomplete extractions, and consider adding a plain-text fallback field on the form.

FAQ

Can I build this without coding?

Yes. The entire workflow is configured through visual tools and form fields. The most technical part is writing the Gemini prompt that defines how CVs are scored, which is plain English text you craft with your client.

How long does it take?

A working version takes around three to four hours to build and test from scratch. Adding a branded Google Sheets dashboard and customising the scoring prompt for a specific role adds another one to two hours depending on how much you iterate with the client.

What can I charge?

Pricing depends on the scope you agree with the client. A basic delivery covering setup, a configured prompt, and a branded dashboard is a defined deliverable. Ongoing retainers for prompt updates, role changes, or adding new positions are a natural upsell once the client sees the system working.

Which tool is required vs optional?

Google Sheets and Google Gemini are core to the workflow as described. Mistral handles PDF extraction and is required unless you substitute a different OCR service. The intake form is built into n8n, so no third-party form tool is needed unless the client already has one they want to use.

This is original DigiNo analysis. The underlying automation pattern is a community workflow template – view the original on n8n.