n8n AI Resume Screening Workflow with a Person Making Every Decision
This n8n workflow gives you an application form with PDF CV upload. Each CV is extracted to text, saved to Google Drive and compared with the role's criteria by an OpenAI or Anthropic chat model, which returns a match level and a short reason for each criterion. The result is logged to Google Sheets and emailed to the hiring manager. No candidate is rejected automatically.
- Difficulty
- Intermediate
- Setup time
- About 45 minutes
- Nodes
- 9
- Checked against
- n8n 2.41
- n8n Form
- Google Drive
- OpenAI or Anthropic
- Google Sheets
- Gmail
Last reviewed October 2026
What this workflow does
A single job ad can bring in a large pile of applications, and a hiring manager reading each CV closely while doing their own job is slow. The usual shortcut is skimming, which is where good candidates get missed and where unconscious bias creeps in. Applicant tracking systems help, but many small businesses still take applications by email and track them in a spreadsheet.
This template gives those businesses a structured first read. The AI checks each CV only against criteria you write, quotes its evidence and suggests interview questions for gaps. The hiring manager still reads the CV and records the decision. That keeps the workflow useful as a reading aid while leaving judgement, and accountability, with a person.
Who it's for
- →Small businesses hiring without an applicant tracking system
- →Hiring managers who screen CVs alongside their day job
- →HR teams that want consistent, criteria-based first reads
- →Recruiters who want written reasons alongside every assessment
How it works, step by step
- 1
Application form
Candidates fill in an n8n form with name, email, mobile, the role and a PDF CV. The form text explains that AI assists and a person decides.
- 2
Extract CV text
Extract from File reads the text out of the uploaded PDF and keeps the original file so it can be stored.
- 3
Save the CV to Drive
Google Drive uploads the PDF to a restricted folder and names it with the candidate and role.
- 4
Load role criteria
A Code node looks up the selected role's must-have and nice-to-have criteria and the hiring manager's email, and builds a link to the stored CV.
- 5
AI scores against the criteria
A Basic LLM Chain asks the chat model to assess only the listed criteria, ignore protected attributes, give evidence for each and suggest interview questions.
- 6
Structured result
The Structured Output Parser forces a fixed JSON shape: overall match, summary, a yes, partly, no or unclear rating per criterion with evidence, and questions.
- 7
Log to Google Sheets
Google Sheets appends a row with the candidate details, the AI reading, criteria notes and the CV link, with Decision set to 'Awaiting review'.
- 8
Notify the hiring manager
Gmail sends the manager the summary, per-criterion notes, suggested questions and the CV link, with a reminder that no decision has been made.
What you need
OpenAI or Anthropic API key
For the chat model that reads each CV. Choose a provider whose data handling terms suit candidate data.
Google Drive
A folder that only the hiring team can access, for stored CVs.
Google Sheets
A screening sheet with columns Submitted, Name, Email, Mobile, Role, AI match, AI summary, Criteria notes, CV and Decision.
Gmail
A recruitment mailbox to notify hiring managers.
Set it up in n8n
- 1
Import the workflow and open 'Application Form'. Edit the role options and the description so it links to your privacy policy.
- 2
Connect Google Drive in 'Save CV to Drive' and select a restricted folder.
- 3
In 'Load Role Criteria', add an entry for every role in the form, with job-related must-haves, nice-to-haves and the hiring manager's email.
- 4
Add your OpenAI credential, or replace the model with an Anthropic Chat Model node connected to the chain.
- 5
Create the screening sheet with the listed columns and select it in 'Log to Screening Sheet'.
- 6
Connect Gmail in 'Notify Hiring Manager'.
- 7
Submit a test application with your own CV through the form's test URL and check the Drive file, sheet row and email.
- 8
Activate the workflow and use the production form URL in your job ads.
Ways to customise it
- +Store role criteria in a Google Sheet or Notion database so hiring managers can edit them without opening n8n.
- +Send candidates an acknowledgement email when their application arrives.
- +Accept Word documents by adding a conversion step before text extraction.
- +Write results into your applicant tracking system, such as Greenhouse, through its API.
- +Add a weekly digest of applications per role for the hiring manager.
For Australian businesses
Federal and state anti-discrimination laws, and the Fair Work Act's general protections for prospective employees, apply to how you screen applicants, including when software helps. The workflow keeps a person making every decision and tells the model to ignore protected attributes, but bias can still come through proxies, so review results across roles and candidates. The employee records exemption in the Privacy Act 1988 does not cover job applicants, so APP obligations apply if your business is covered: give a collection notice (APP 5), keep CVs secure (APP 11), and check where your AI provider processes data (APP 8). Set a retention period and delete unsuccessful applications when it ends.
Tools in this workflow
Greenhouse
PaidHR
An applicant tracking and recruiting platform built around structured hiring: job kickoffs, scorecards, interview kits and consistent evaluation. Its AI features include talent matching, spam and fraud detection on applications, identity verification, and assistants that summarise candidate information for hiring teams.
Best for: Growing companies hiring across several teams at once
Greenhouse review →ChatGPT (OpenAI)
Free plan + paidGeneral Assistant
OpenAI's general-purpose AI assistant. It handles writing, coding, file and image analysis, data analysis and web research through a chat interface, with an API for building it into your own systems.
Best for: Small and mid-sized teams wanting one general assistant for everyday writing and analysis
ChatGPT review →Claude (Anthropic)
Free plan + paidGeneral Assistant
Anthropic's AI assistant, known for careful reasoning, long-document analysis, writing and coding. It can work through large documents and codebases in a single conversation, paid plans include the Claude Code agentic coding tool, and it is available through an API and major cloud platforms.
Best for: Legal, policy and consulting teams reviewing long documents
Claude review →n8n
Free plan + paidAutomation
A workflow automation platform that combines a visual node editor with code when you need it. It connects hundreds of apps and APIs and includes AI agent nodes that can call language models, tools and memory. Teams can self-host the free Community Edition on their own servers or use n8n's hosted cloud.
Best for: Australian organisations that need automation data to stay onshore
n8n review →
Frequently asked questions
Does this workflow reject candidates automatically?
No. Every application is logged with Decision set to 'Awaiting review' and sent to the hiring manager. The AI gives a reading against your criteria; the manager reads the CV and decides. Adding automatic rejections would increase legal and fairness risk.
How does it reduce bias?
The prompt limits the model to job-related criteria you write and tells it to ignore name, age, gender, address, photos and other protected attributes. That helps, but it cannot remove bias on its own, so a person reviews every application and you should check outcomes regularly.
What file types can candidates upload?
The form accepts PDF, which Extract from File reads directly. Scanned image PDFs contain no text, so the AI will mark them as unclear and the manager reads them normally. You can add OCR or Word conversion if you need it.
Can I use Claude instead of OpenAI?
Yes. Add an Anthropic Chat Model node, connect it to the 'Score CV Against Criteria' chain and delete the OpenAI node. The structured output parser works with either provider.
Do privacy laws apply to job applicants?
Yes, for businesses covered by the Privacy Act. The employee records exemption applies to employees, not applicants, so you need a collection notice, secure storage and a sensible retention period for CVs and screening notes.
More operations & hr templates
Other n8n templates
- Speed to Lead: AI Lead Qualification from Website Form to HubSpot
- AI Company Research Agent: Lead Research Briefs Saved to HubSpot
- AI Email Triage for Gmail: Labels, Draft Replies and a Daily Digest
- RAG Website Support Chatbot from Your Google Drive Docs with Human Handoff
- Supplier Invoice Email to Xero Draft Bill with AI Extraction
- Tiered Overdue Invoice Reminders from Xero
Want this running in your business?
AI Lab Australia sets up n8n workflows end to end: hosting in Australia, connecting your systems, testing with real data and handing over documentation.