UbertAI engineering for SMEsBook an intro call
Feature · HR & recruitment

Forty CVs in the inbox and nobody with time to read them. Candidates hear nothing for two weeks, and go to the competitor next door.

HR in an SME is an inbox full of CVs, an Excel sheet with leave balances and a statutory sick-leave deadline nobody keeps an eye on. AI can do a lot here, but recruitment is also where the law is strictest: selection by algorithms becomes high-risk, and automatic rejection is already prohibited. So we draw a sharp line: AI for the work around it, people for the judgement.

// A feature at a fictional logistics company with 45 employees, one HR officer who also prepares payroll, three open vacancies and a managing director who conducts the job interviews himself. The company is fictional, the situations are real: we run into them again and again.

Book an intro call Take the free AI scan

Sound familiar?

  • Forty CVs in the owner’s inbox; candidates hear nothing for two weeks.
  • The 2019 vacancy text with “young, dynamic colleague”. Indirect age discrimination, and nobody notices.
  • Six emails and three chat messages to find one interview slot.
  • The cover letter that was clearly written by ChatGPT, and nobody knows what it is worth anymore.
  • The new colleague spends week one waiting for a laptop and login details.
  • Sick leave reported to the manager via WhatsApp; nobody enters it in the system and the statutory sick-leave deadlines keep running.
  • Leave in Excel; the balance does not match at payroll time.
  • Leavers: laptop handed in, but Drive, Slack and the webshop admin are still open.

What usually runs here

The software we come across here most often. It does not have to go; we build around it and in between.

AFAS Profit HR, payroll, ERPNmbrs payroll via the accountantLoket.nl payroll and HRHoorayHR / Employes leave, absence, personnel filePersonio HR suite with ATSRecruitee Dutch ATSHomerun entry-level ATSIndeed / LinkedIn / Nationale Vacaturebank vacanciesLinkedIn Recruiter active sourcingArboNed / HumanCapitalCare absence, statutory sick-leave processOutlook / Google Calendar scheduling interviews
// the same working day, twice

A day in hr & recruitment: now and next

On the left, the day as it usually runs now. On the right, the same day once the dull parts are automated and AI does the groundwork. Watch what changes: not the people, but where their time goes.

// how it goes nowWithout AI
  1. 08:00HRInbox: 23 new applications for three vacancies. Open, read, drag into a folder. The first five candidates from last week are still waiting for a reply.
  2. 09:30Director“Can you set up an interview with that one from Zwolle?” HR emails the candidate. Reply on Tuesday. The director didn’t check his inbox on Tuesday.
  3. 11:00HRA team leader reports that an employee has been off sick since Monday. Via WhatsApp. HR enters it in the system, two days late. Week 6 is approaching.
  4. 13:00HRNew colleague starts tomorrow. The laptop hasn’t been ordered yet; she’ll ask IT for the accounts later. IT isn’t in tomorrow.
  5. 14:30HRThe leave balance in Excel doesn’t match the payroll run. An hour spent looking for the difference.
  6. 16:00DirectorInterview with a candidate. No preparation, no question list, a good feeling. Next week another good feeling.
  7. 17:30HRAn employee who left last month turns out to still have access to the shared Drive. Nobody had a checklist.
// how it will goWith AI and automation
  1. 08:00HRThe 23 applications have been received and confirmed with a personal message and a timeline. A neutral summary is ready for each candidate: experience, availability, what is missing. No score, no ranking; HR does that itself.
  2. 09:30DirectorThe candidate from Zwolle picked an interview slot from the director’s calendar themselves. The preparation sheet with standard questions is ready.
  3. 11:00HRThe sick-leave report came in via the app, is in the system and the statutory sick-leave deadlines are scheduled with reminders. The team leader has the conversation with the employee.
  4. 13:00HR“Contract signed” set the laptop, accounts, badge and onboarding checklist in motion yesterday. Tomorrow everything is ready.
  5. 14:30HRLeave and absence are in one system that feeds the payroll run. No Excel, no difference.
  6. 16:00DirectorInterview with the candidate using the same questions as for the others; notes in the file. Comparing is now fair and fast.
  7. 17:30HRThe departed employee is logged out everywhere with one click: mail, Drive, Slack, webshop. The log shows what was closed.

Where AI makes the difference here

Per use case: what the automation does, what a person keeps doing, and whether it is proven or still promising. We say which one honestly.

01proven

Vacancy texts without exclusionary language

AI writes a draft in your tone, flags exclusionary language (age, gender, jargon) and creates variants for Indeed and LinkedIn.

Still human: sets the requirements and the tone.

Working more efficiently
02proven

Candidates who always get a reply

Confirmation, status update and a courteous rejection as a draft, on time. A chatbot for candidates identifies itself as AI.

Still human: sends the rejection and conducts every interview.

Saving time
03proven

CVs summarised, not scored

A neutral summary per applicant: experience, education, availability, what is missing. Deliberately no ranking: that becomes high-risk, and automatic rejection is already prohibited.

Still human: ranks and decides, with the same questions for everyone.

Working more efficiently
04proven

Scheduling interviews without email rounds

The candidate picks a slot from linked calendars; preparation and question list are ready.

Still human: conducts the interview.

Saving time
05proven

Onboarding and offboarding as one button

“Contract signed” sets laptop, accounts and checklist in motion; “left the company” closes all access the same day, with a log.

Still human: approves exceptions and welcomes the new colleague in person.

Better security Working more efficiently
06proven

Absence and statutory sick-leave deadlines monitored

Deadlines (week 6, 8 and 42) automatically scheduled with reminders; the file is complete for the company doctor.

Still human: has the conversations; that stays human work.

Saving time
83%
of former employees could still log in at their old employer after leaving. [1]
57%
of applicants use AI when applying; 40% of employers do. [2]
26%
of candidates trust that AI will evaluate them fairly. [3]

Honestly: what AI does not do here

  • AI mustn’t decide who gets rejected. A “human review” that simply rubber-stamps every AI score doesn’t count as a human decision. So no scores and no rankings from a machine.
  • Selection by AI becomes high-risk under the AI Act (currently planned for December 2027). Anyone who wants to go down that path must implement the full set of safeguards. For SMEs we advise against it.
  • The interview, the rapport and the decision are and remain human. AI only takes out the waiting and the typing.
  • Rosters, appraisals and systems that monitor employees require the consent of the works council. We arrange that beforehand, not afterwards.

The objections we hear — and whether we can close them off

We looked them up and heard them from clients. For each objection we say whether it is really solvable, partly solvable, or a risk that stays and that you accept knowingly.

Can it be closed off: partly. That risk is real and cannot be engineered away; the employer remains liable, not the supplier. That’s why we don’t let AI rank or reject candidates. What AI does do: remove exclusionary language from vacancies, make sure everyone gets the same questions and record the process. That makes selection fairer, not less fair.

// how we start, the first thirty days
  1. 01Week 1: core session with HR and management; we follow one vacancy from text to first working day and one sick-leave report from chat message to file.
  2. 02Weeks 1 and 2: candidate communication and interview scheduling, linked to your ATS or inbox; vacancy texts with a language check.
  3. 03Weeks 2 and 3: onboarding and offboarding flow with account management and a log.
  4. 04Weeks 3 and 4: absence registration with statutory sick-leave deadlines; leave and absence into one source for the payroll run.
  5. 05Week 4: inform the works council and arrange consent where needed; measure: response time to candidates, lead time, open access after leaving (must be zero).
// what it costs and how it works

First the free AI scan, then a core session on your floor (a half-day of 4 hours, €596 excl. VAT and travel costs) with a core report and an honest go/no-go. We build on a project basis with fixed hours: €110 per hour for straightforward work, €165 for complex development.

More hours only with your written approval. If we cannot deliver, you pay nothing for what was not delivered.

Process automation AI guidance & training AI compliance & grants

Want to know what this means for your hr & recruitment?

Take the free AI scan or book an intro call with someone who builds it themselves. A reply within one working day, no slides, no obligations.

Book an intro call Take the free AI scan
// which industries this matters most in
Professional services Healthcare & social care
// from our insights
Where do you start with AI in your business?

Most businesses start with the tool and then get stuck. It is better to start with one piece of work that comes back every week and nobody enjoys.

What does AI automation really cost?

The build costs are usually the easiest part of the bill. The surprises are in the clean-up beforehand, the monthly usage and the maintenance afterwards.

// other departments
Administration & finance Sales & quotes Purchasing & stock Production & workshop Construction & installation Field service & maintenance Transport planning & administration Marketing & content Customer service & communication Planning, scheduling & events IT, security & data Management & leadership
// sources
  1. Beyond Identity, Former employees admit to using continued account access (2022)
  2. Werf& (Indeed), Employers lag behind in AI use (2025)
  3. Gartner, Just 26% of job applicants trust AI will fairly evaluate them (2025)
  4. Greenhouse, 63% of job seekers have faced an AI interview (2026)
  5. Netherlands Institute for Human Rights, Almost all employers use algorithms for job applications (2022)
  6. AI Compliance Hub, summary of the Dutch DPA report AI & Algorithms (RAN-6) (2026)
  7. Dutch Data Protection Authority, Job applications (retention periods) (2025)
  8. EUR-Lex, Regulation (EU) 2026/1744 (Digital Omnibus on AI) (2026)
  9. SBI Formaat, Works Councils Act article 27: works council right of consent (2025)
  10. HR Praktijk (Personio/Censuswide), Risk of an HR brain drain at Dutch companies (2025)
  11. World Economic Forum, Future of Jobs Report 2025 (2025)