AI Recruiter Zurich Gipfeli working when humans sleep


As you read this, there could be a a tech start-up in Zurich, Basel or a group of Budapest or Mumbai based software engineers working on new methods to change the way candidates are found, sorted, and recruited.

Not just for the low level high turnover roles, they are working on about every job title imaginable, including senior management. The idea that our friend pictured above will replace me does not present me with too many sleepless nights just yet.

Human resources professionals will become displaced by hiring robots, yet at the same time the fact that many think their next job depends on the accuracy of a complex, yet intentionally unbiased 24/7 operating algorithm, feeds into a movie type narrative.

Should younger managers leave HR and plan to be made redundant by our friend here? Well he does stand as if he has your job already, but you are too young to remember dial-up, the job boards being replaced globally by Linkedin and a lot more.

Companies like ABB or E&Y give Linkedin far more than CHF 500,000 globally and yet they still employ internal and external recruiters. So what are the trends and how will they kill off recruiters? To understand my future and certain passing because of AI, I think we need to look at a few of the daily or process related activities that recruiters and hiring managers are inolved with when sourcing and employing new staff.


Online chat and video candidate qualification – Robo-chat and robots replace recruiters

This will be a big disruptor, analysis by video link and messaging post could be another. There is no need to panic, Implenia will not adopt such technology to find its construction staff from Vaud, Valais or Appenzell, the accents will kill any current system outright. Recruitment is a process and many steps exist today as they always have done so, yet trials of robo-processing by banks have come and some shelved. The talk is good but the reality is different, the technology and development is some way off getting past very basic repetitive functional work.

Resume sorting and selecting

Resume scanning is time consuming, in the old days it was a manual paper process and then many systems developed tools for keyword search and document validation. In 2004 many companies in Switzerland wanted paper copies, very little was electronic. The development of applicant tracking system (ATS) software together with its keyword search tools, tags, database management and intelligent rating means hundreds of pages are processed in seconds. In truth when you have 200 applications using “apply with Linkedin,” it makes sense, yet volume presents problems as someone has to check the list, uncover the badly written CV by the “high potential” or execute a compliance and quality check. The human involvement may be less, however in the future it could be to work less concentrating more on narrowing the qualified candidate selection number for interview. This is especially important when certain scientific or engineering jobs require a deep dive understanding of a topic which may go beyond the limits of a AI logorithm or basic keyword search.

AI is young and handicapped by its advocates

HireVue is a company which markets software for screening job candidates who sit in front of a camera and are asked a few questions. Nathan Mondragon, head psychologist for HireVue was quoted in the Guardian newspaper as saying “the camera delivers images to the candidate that takes note of barely perceptible changes in posture, facial expression and vocal tone. We break the answers people give down into many thousands of data points, into verbal and non-verbal cues. If you are answering a question about how you would spend a million dollars, your eyes would tend to shift upwards, your verbal cues would go silent. Your head would tilt slightly upward with your eyes.”

I will leave you to digest this service offer as being of any value in an interview process today, the concept is great but any current hiring manager or recruiter will want to look the person in the eye and make their own conclusion based on more than one question. HireVue will morph into something else in the future, then I would expect it to return with more than just this current commercial application.

Eliminating Bias

AI is touted as being the key to promoting diversity and eliminating bias, does this really have a chance to be applied in Switzerland? Ageism in hiring is well documented, news articles are published every quarter and there are so few women CEOs in companies, for this market many are talking of AI as a “pretend technology adoption.” Human bias factors can still enter into any hiring decision from the moment of the first phone call conversation or in-person interview. Women in business at C-level are less than 10% of the total in many top companies, adding AI to the hiring process will not change that culture or percentage soon.

Where it could possibly work would be where skills are easy to define, to match education and experience. That is only a small part of the total picture of hiring, onboarding and employee or executive development.

What happens if someone writes the wrong code and unintentionally does not hire the right candidates, so discriminates against others by mistake? Amazon were victim of this using AI testing in their recruitment process in 2014. Within 12 months of the initial test phase starting, it had to be stopped because it was downgrading women applicants against men. Reuters exposed the project faults and the company search for the hiring AI holy grail was downgraded.

AI has a long way to go, the AI-based hiring assessment tools are close to the bottom of the technology adoption curve and AI-based assessment innovation is costly. Talk to any HR or Learning and Development professional and they can give you timelines and in depth analysis, in areas where AI has yet to have any meaningful investment. This gap to use abstract reasoning to perform complex-judgment-based tasks and to understand deeper levels of meaning from the data is just not robust enough to be compared to the tried and tested models for psychometric testing.

AI – millions and millions of data points which not recruiter insight, intuition or experience.

No-one who has been in recruitment or hiring has had 100% success at hiring the new employee or key manager, speak to anyone in HR and you will get some interesting stories where “we got it wrong.” That past experience and learning has yet to be properly documented, to have an algorithm written about each event or decision and its outcomes or communication flow. That is why I am sleeping well at the moment, saying to young HR managers, “you have at least 15-20 years.”

Human beings are complex, no AI software, no matter how well engineered will be able to tell pre-empt the usefulness of HR or the recruiter who knows which candidate is moving because she is a huge Davos or Kloten ice hockey fan, or that the new CEO who relocated the office from Basel to Zug did so because of his passion for rock climbing and paragliding.

AI will make recruiters more reliant on technology, work methods will change, it is a certain, yet thankfully his lordship at the top of the page still has some waiting to do!

Author – Andrew Travers


Leave a reply

Your email address will not be published. Required fields are marked *


three × 2 =

©2019 Designed by Travers Business Development