AI automation for recruitment agencies

CVs land in your CRM formatted and anonymized, before you have opened your inbox.

Your morning starts with one email, not forty. Built for solo recruiters and agencies up to about twenty people. Live in four weeks.

A recruiter’s morning, before and after.

Now
  • Open inbox: 40+ emails, CVs buried in threads
  • Download each CV, rename the file, upload to the ATS
  • Retype candidate data the parser missed
  • Reformat two CVs into the agency template by hand
  • Delete names and contacts before sending to a client, hope nothing slipped
  • Realise at noon a strong candidate applied on Friday
With the system
  • 7:30, one digest email: what came in, what matters, what is on today
  • Every overnight CV already parsed, formatted, in the ATS
  • Client-ready anonymized versions waiting for your check
  • A dormant candidate re-engaged himself into your pipeline
  • Low-confidence parses flagged for a human, not guessed
  • You spend the morning talking to candidates
The package

Recruiter Autopilot: four weeks, four systems.

Not “custom automation services”. A named build with a fixed scope, delivered week by week on n8n and Claude, connected to the ATS you already run.

  • 1
    Week 1

    CV intake

    Email attachments become parsed, structured ATS records automatically. Works with the inbox you already have.

  • 2
    Week 2

    Morning digest

    One 7:30 email: what came in overnight, what needs attention, what is on today. Inbox archaeology ends.

  • 3
    Week 3

    CV formatting & anonymization

    Every CV lands in your template, anonymized for client sending, with a human check before anything goes out.

  • 4
    Week 4

    Re-engagement & lead-gen

    Your dormant candidate database starts working again: matched, messaged, and routed back to your desk.

  • Price and terms

    Fixed price, agreed in writing before work starts. Delivered in four weeks. Optional monthly support: monitoring, alerts and fixes, not billable hours. One call gets you the exact number.

    Inside the system

    What it actually looks like.

    Client work is confidential, so these are faithful recreations with sample data, the same shape as the production system, none of it from a real candidate.

    Sample data

    Your desk this morning · Tue 7:30

    • 4 new CVs overnight, parsed and in the ATS. 2 match the Warsaw .NET role.
    • 1 reply from a re-engaged candidate (dormant 14 months): open to contract work.
    • 2 interviews today, confirmations received for both.
    • 1 attention item: CV from J. K. arrived as a photo scan, parsed with low confidence, review before sending.
    Before · sample

    MARTA W. · marta.w.1990@gmail.com · +48 600 000 000

    senior java developer, 8 yrs exp, warsaw, looking for b2b...

    CV_final_v3_NEW(2).docx · 4 fonts · photo included · client sees name, contacts, current employer

    After · sample

    Candidate MW-2481 · Senior Java Developer

    8 years · Warsaw · available from March · B2B

    Agency template · anonymized · one font · skills mapped to the role spec · ready to send

    Case in production

    A solo recruiter, CEE market: the system this page describes is her desk.

    Stack: Gmail, Google Drive, Manatal, n8n, Claude. What hurt: CVs arriving by email and sitting there, every CV reformatted and anonymized by hand, mornings lost to inbox triage. What runs now: automated CV intake into Manatal, the 7:30 digest, formatting and anonymization with human review. Re-engagement and lead-gen are the current build week. Measured numbers land here as the case matures; we publish what we can verify, nothing else.

    Before this: nine years of CRM and automation engineering, including a recruitment ATS we built and shipped on the Salesforce AppExchange.

    The light option

    12 things recruiters still do by hand, and the 4 worth automating first.

    Not ready for a call? Fair. Leave an email, get the checklist right here, no PDF attachment, no drip campaign. Prefer a look at your own process? Send it through the contact form and you get a 5-minute Loom breakdown back.

    The questions recruiters actually ask.

    Does this work with Manatal, Bullhorn, Vincere or Recruit CRM?

    Yes. The current production build runs on Manatal; the same pattern connects to Bullhorn, Vincere, Recruit CRM, Loxo or JobAdder through their APIs. The workflow is the product, the ATS is a connector.

    Is this safe for my LinkedIn account?

    Honest answer: mass LinkedIn automation from cloud tools is against LinkedIn’s terms and risks your account. Where LinkedIn is involved we keep actions in your own browser, at human pace, with you pressing the final button. Most of the value here (CV intake, formatting, digest, re-engagement) does not touch LinkedIn at all.

    What happens when it breaks at 7am?

    Every workflow ships with monitoring and error alerts. A failed step becomes a message to us, not a silent gap in your morning digest. Fixing what breaks is part of the support arrangement, not an hourly invoice.

    Who owns the workflows after handover?

    You do. Everything runs in accounts you control, documented, with a written spec. If we disappear tomorrow, your system keeps running and any competent automation person can maintain it.

    How much does it cost?

    Fixed price for the four-week build, agreed in writing before work starts, plus an optional monthly support arrangement for monitoring and fixes. The exact number depends on your stack and volume; you will have it after one call, not after the project.

    Stanislav Shaban

    I’m Stan. I build these myself.

    No account managers, no juniors. Nine years of automation and CRM engineering, a QA background before that, and one production recruitment system on my desk right now. On the call you talk to the person who does the work.