Recrutei + OpenAI: searchable CV notes, no rework

CVs arrive. Names get copied. Vacancies get mis-selected. Then someone spends their afternoon fixing a pile of “almost right” candidate records that should’ve been clean from the start.

This is where recruiters lose time, honestly. HR managers feel it when pipelines get messy and reporting stops making sense. And if you run a small agency, you can’t afford rework. This Recrutei OpenAI automation routes applicants to the right vacancy and turns the PDF resume into searchable notes inside your ATS.

Below you’ll see how the workflow runs, what it changes day-to-day, and what you need to implement it without turning your hiring process into a tech project.

How This Automation Works

See how this solves the problem:

n8n Workflow Template: Recrutei + OpenAI: searchable CV notes, no rework

Click to explore

flowchart LR

    subgraph sg0["On Form Submission Flow"]
        direction LR
        n0["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/form.svg' width='40' height='40' /></div><br/>On Form Submission"]
        n1["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Create Prospect in Recrutei"]
        n2["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Get Pipe Stages"]
        n3@{ icon: "mdi:robot", form: "rounded", label: "Get Vacancy ID from AI", pos: "b", h: 48 }
        n4@{ icon: "mdi:swap-vertical", form: "rounded", label: "Set Vacancy ID", pos: "b", h: 48 }
        n5@{ icon: "mdi:wrench", form: "rounded", label: "Recrutei's MCP", pos: "b", h: 48 }
        n6["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Add Curriculum as Observation"]
        n7@{ icon: "mdi:cog", form: "rounded", label: "Extract Text from PDF Resume", pos: "b", h: 48 }
        n8["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/merge.svg' width='40' height='40' /></div><br/>Merge Candidate Data"]
        n5 -.-> n3
        n4 --> n2
        n2 --> n1
        n0 --> n3
        n0 --> n8
        n8 --> n7
        n3 --> n4
        n1 --> n8
        n7 --> n6
    end

    %% Styling
    classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
    classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
    classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
    classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
    classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
    classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
    classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
    classDef disabled stroke-dasharray: 5 5,opacity: 0.5
    class n0 trigger
    class n3 ai
    class n5 ai
    class n1,n2,n6 api
    classDef customIcon fill:none,stroke:none
    class n0,n1,n2,n6,n8 customIcon

The Challenge: CV intake that creates rework

Most “Apply Now” funnels look simple until you’re the person cleaning up the output. Candidates type a vacancy name that doesn’t match what’s in your ATS. Someone picks the wrong opening because two job titles sound similar. The resume sits in a file field, so you can’t search it later, and the same CV gets opened and reread again and again. It’s not just slow. It’s mentally noisy, and small mistakes compound when you have multiple roles open at once.

It adds up fast. Here’s where it breaks down in real teams.

Applicants select “Javascript Developer” while your ATS uses “Front-end Engineer,” so the candidate lands in the wrong vacancy or gets stuck waiting for manual triage.
Someone has to fetch pipe stages, choose the right first stage, and create a prospect by hand, which is easy to do wrong when you’re rushing.
Resume content stays trapped in a PDF upload, which means you can’t search for “React,” “Spanish,” or “B2B SaaS” inside Recrutei when you need a quick shortlist.
When the ATS record is incomplete, the team fills the gaps in Slack, email, and spreadsheets, and nobody trusts the pipeline view anymore.

The Fix: form-to-Recrutei intake with AI vacancy matching

This workflow turns your application form into a clean, consistent intake channel for Recrutei. A candidate submits an n8n form with their contact details, the vacancy name they chose, and a PDF resume. OpenAI then helps map that human-friendly vacancy label to the exact vacancy_id Recrutei expects, so you aren’t relying on someone to guess correctly. Once the workflow has the right vacancy, it fetches the pipe stages for that role, creates the prospect in the first available stage, and merges everything back together so the resume is still attached to the same candidate. Finally, it extracts the text from the PDF and posts it into the candidate profile as an internal observation, making the resume searchable right inside Recrutei.

The workflow starts at the form submission, then AI resolves the correct vacancy ID, and Recrutei receives a properly staged prospect. After that, the PDF is parsed and the CV text is saved as a note, so search works the way recruiters expect it to.

What Changes: Before vs. After

What This Eliminates

Impact You’ll See

Manually looking up the correct vacancy when job titles don’t match.
Creating prospects one-by-one and guessing the right pipe stage under pressure.
Copying resume text into notes (or not doing it at all).
Re-opening PDFs repeatedly because the ATS can’t search the content.

Most teams save about 5 hours a week once volume picks up.
Candidates land on the right vacancy far more consistently, so you stop doing “routing cleanup” at the end of the day.
Shortlisting gets faster because “search in Recrutei” starts working like a real filter, not a guessing game.
Less back-and-forth between spreadsheet trackers and the ATS because the record is complete at creation.
A more reliable funnel metric story, since prospects are created with consistent vacancy and stage data.

Real-World Impact

Say you receive 20 applications a week across 4 open roles. Manually, it’s common to spend about 10 minutes per application finding the right vacancy, creating the prospect, choosing the stage, and adding a quick resume note (that’s roughly 3 hours a week). With this workflow, the submission takes the candidate a minute, and your team mostly just reviews what’s already in Recrutei. You get those hours back, and the “wrong vacancy” problem largely disappears.

Requirements

n8n instance (try n8n Cloud free)
Self-hosting option if you prefer (Hostinger works well)
Recrutei to create prospects and attach observations.
OpenAI to match vacancy name to vacancy_id.
Recrutei API key (get it from Recrutei token docs/admin).

Skill level: Intermediate. You’ll connect accounts, add API credentials, and test with a real PDF resume.

Need help implementing this? Talk to an automation expert (free 15-minute consultation).

The Workflow Flow

Application submission triggers the run. A candidate completes your n8n form with name, email, phone, the vacancy name they chose, and a PDF resume.

AI resolves the vacancy_id. The vacancy name from the form is sent to an AI Agent powered by OpenAI, which uses a Recrutei MCP tool connection to find the official vacancy ID your ATS expects.

Recrutei prospect creation happens with the right stage. The workflow fetches pipe stages for that vacancy, then creates the prospect via the Recrutei API using the first available stage so the candidate appears in the correct pipeline immediately.

The resume becomes searchable inside the ATS. n8n extracts text from the PDF and posts it back to the candidate profile as an internal observation (note), so recruiters can search and filter using real CV content.

You can easily modify the form fields and the AI matching prompt to fit your naming conventions and screening rules based on your needs. See the full implementation guide below for customization options.

Step-by-Step Implementation Guide

Step 1: Configure the Form Trigger

Set up the intake form that starts the workflow and captures candidate details.

Open Form Intake Trigger and set Form Title to Job Application Form.
Set Form Description to Apply for the job of your choice..
Confirm the form fields match the workflow: Full Name, Email, Phone, Vacancy (dropdown), and Resume (file).
Verify the submit button text under Options is Submit and the success message is Your application has been successfully submitted..

Form Intake Trigger outputs to both AI Vacancy Lookup and Combine Applicant Data in parallel.

The Flowpast Branding note is informational only and does not affect execution.

Step 2: Connect Recrutei API Requests

Configure the Recrutei HTTP requests to fetch pipeline data and create the prospect record.

In Retrieve Pipe Stages, set URL to =https://api.recrutei.com.br/api/v1/vacancies/{{ $json.id }}/pipes and keep Send Headers enabled.
In Retrieve Pipe Stages headers, replace Bearer [CONFIGURE_YOUR_TOKEN] with your Recrutei API token.
In Create Recrutei Prospect, set URL to https://api.recrutei.com.br/api/v2/prospects, Method to POST, and Content Type to multipart-form-data.
Map body parameters in Create Recrutei Prospect exactly as shown: name ={{ $('Form Intake Trigger').item.json['Full Name'] }}, pipe_stage_id ={{ $json.id }}, vacancy_id ={{ $('Assign Vacancy Identifier').item.json.id }}, email ={{ $('Form Intake Trigger').item.json.Email }}, telephone ={{ $('Form Intake Trigger').item.json.Phone }}, and prospect_acquisition_channel_id 8.
In Create Recrutei Prospect headers, replace Bearer [CONFIGURE_YOUR_TOKEN] with your Recrutei API token and keep Content-Type as multipart/form-data.

⚠️ Common Pitfall: Leaving Bearer [CONFIGURE_YOUR_TOKEN] unchanged will cause authorization failures in Retrieve Pipe Stages, Create Recrutei Prospect, and Attach Resume Observation.

Step 3: Set Up AI Vacancy Matching

Configure the AI model to translate the applicant’s vacancy selection into a Recrutei vacancy ID.

Open AI Vacancy Lookup and confirm Model is gpt-4.1-mini.
Set the user message content to =Vacancy: {{ $json.Vacancy }}.
Keep the system message text exactly as provided so the model returns only the vacancy ID.
In Assign Vacancy Identifier, set the assignment id to ={{ $json.message.content }} with type number.
Configure Recrutei MCP Client with your MCP endpoint: set Endpoint URL to your value instead of YOUR_MCP_ENDPOINT_URL_HERE.

Credential Required: Connect your OpenAI credentials in AI Vacancy Lookup. The Recrutei MCP Client is an AI tool sub-node connected to AI Vacancy Lookup, so credentials should be added to the parent node.

If the model returns non-numeric output, Assign Vacancy Identifier will fail because id is typed as a number.

Step 4: Merge Applicant Data and Parse the Resume

Combine the form submission with the created prospect and extract text from the resume file.

In Combine Applicant Data, set Mode to chooseBranch and Use Data Of Input to 2 so it uses the prospect creation output.
Configure Parse Resume PDF with Operation set to pdf and Binary Property Name set to Resume.

Step 5: Add the Resume Observation to Recrutei

Send the extracted resume text to Recrutei as a talent observation linked to the application.

In Attach Resume Observation, set URL to https://api.recrutei.com.br/api/v1/talents-observations and Method to POST.
Set Content Type to multipart-form-data and enable Send Headers.
Map body parameters: application_id ={{ $('Create Recrutei Prospect').first().json.data.application.id }}, talent_id ={{ $('Create Recrutei Prospect').first().json.data.talent.id }}, talent_observation_type_id 11, and description ={{ $json.text }}.
Replace the Authorization header value Bearer [CONFIGURE_YOUR_TOKEN] with your Recrutei API token.

⚠️ Common Pitfall: If the resume upload field is missing or the file is not a PDF, Parse Resume PDF will not produce $json.text, and Attach Resume Observation will send an empty description.

Step 6: Test and Activate Your Workflow

Run a full test to validate AI matching, Recrutei creation, and resume parsing before enabling production use.

Click Execute Workflow and submit a test entry via Form Intake Trigger with a PDF resume and a valid vacancy selection.
Verify AI Vacancy Lookup returns a numeric ID and Assign Vacancy Identifier sets id correctly.
Confirm Retrieve Pipe Stages and Create Recrutei Prospect return successful responses from Recrutei.
Check that Attach Resume Observation creates an observation containing parsed resume text.
When satisfied, toggle the workflow to Active so it runs on every new form submission.

🔒

Unlock Full Step-by-Step Guide

Get the complete implementation guide + downloadable template

Watch Out For

Recrutei credentials can expire or need specific permissions. If things break, check the Authorization bearer token in your n8n credential and confirm the Recrutei API access still works.
PDF resumes aren’t always clean text. Some are scanned images, which means extraction can return very little. If your notes look empty, test with a different PDF and consider adding OCR upstream.
OpenAI vacancy matching depends on naming consistency. If your team renames vacancies often, update the system prompt (and confirm the MCP endpoint URL) so the AI keeps mapping to the right vacancy_id.

Common Questions

How quickly can I implement this Recrutei OpenAI automation?

About an hour if your Recrutei API access is ready.

Can non-technical teams implement this vacancy matching automation?

Yes, but someone needs to be comfortable pasting API keys and testing a few real submissions. No coding, just careful setup.

Is n8n free to use for this Recrutei OpenAI automation workflow?

Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in OpenAI API costs, which are usually a few cents per batch of applications.

Where can I host n8n to run this Recrutei OpenAI automation?

Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.

How do I adapt this Recrutei OpenAI automation solution to my specific challenges?

Start with the AI Vacancy Lookup prompt, because that’s where your naming quirks show up. You can tighten matching by adding examples of “messy” vacancy names your candidates use, plus the exact titles that exist in Recrutei. If your process needs a different starting stage, swap the “first available pipe stage” logic in the Retrieve Pipe Stages and Create Recrutei Prospect steps. Many teams also add extra form fields (location, salary expectation, portfolio URL) and merge them into the prospect payload.

Why is my Recrutei connection failing in this Recrutei OpenAI automation workflow?

Usually it’s an expired or wrong bearer token in the n8n credential used by the HTTP Request nodes. Update the Authorization header, then retest the “Retrieve Pipe Stages” call first since it’s the quickest signal. If that works, check that your MCP endpoint URL is correct and reachable. Rate limits can also show up when you test with a burst of submissions, so space out runs during troubleshooting.

What’s the capacity of this Recrutei OpenAI automation solution?

On n8n Cloud, capacity depends on your plan’s monthly executions; Starter covers a typical small hiring team, and you can move up as volume grows. If you self-host, there’s no execution limit from n8n itself, but your server and the Recrutei/OpenAI APIs become the practical bottlenecks. In real use, this workflow is fine for day-to-day inbound applications, even with multiple roles open at once. If you’re processing hundreds of CVs in a short window, add queueing and retry logic.

Is this Recrutei OpenAI automation better than using Zapier or Make?

Often, yes. This workflow leans on AI + a tool call (the MCP client) and multi-step API handling, which is doable in Zapier/Make but tends to get clunky and expensive as you add logic. n8n also gives you a clean self-hosting path, which matters if you want predictable costs at higher volume. If your intake is truly simple (one form, one “create record” action), Zapier can be faster to set up. But for vacancy matching plus PDF text extraction, n8n is usually the more comfortable long-term choice. If you want a second opinion, Talk to an automation expert and we’ll sanity-check your stack.

Once this is running, your ATS starts feeling dependable again. The workflow handles the repetitive intake work, and your team can stay focused on interviewing and decisions.

Recrutei + OpenAI: searchable CV notes, no rework

How This Automation Works

n8n Workflow Template: Recrutei + OpenAI: searchable CV notes, no rework

The Challenge: CV intake that creates rework