Understanding hiring algorithms: Career coaching in 2019

Today, hiring technology vendors increasingly build predictive features into tools that are used throughout the hiring process.They rely on machine learning techniques, where computers detect patterns in existing data (called training data) to build models that forecast future outcomes in the form of different kinds of scores and rankings.”  Help Wanted: An Examination of Hiring Algorithms, Equity, and Bias

The hiring process is changing faster than most realize. Automated tools that use machine learning to source, analyze, and rank candidates are already integrated into the hiring process, sometimes without candidates knowing it. As we move into 2019, the adoption of predictive hiring tools show no sign of slowing down.

Career coaches and university career services departments have a responsibility to understand these hiring algorithms and their impact on job seekers. They need to create new strategies and update career workshops to help job seekers navigate hiring algorithms. (spoiler alert: my career workshops cover this)

If you don’t work in tech, artificial intelligence and machine learning can seem like intimidating topics. Even more unhelpful, hiring algorithms are a bit of a black box. The transformation to an automated hiring process is happening behind the scenes. It’s hard to figure out which companies use this technology and exactly these tools work. It’s challenging to know which automated tools use questionable data or cement bias into the hiring process.

Luckily, there’s a new report out to help career coaches get up to speed on new HR technology and hiring algorithms. The report, Help Wanted: An Examination of Hiring Algorithms, Equity, and Bias,is produced by the nonprofit Upturn. Upturn’s mission is to promote “equity and justice in the design, governance, and use of digital technology.” Their mission shines through in their new report.

The report offers career coaches a comprehensive review of new hiring technology. The report is well-written, making it an accessible read. More importantly it isn’t filled with marketing promises that dominate HR tech press. Instead, it dives deep into the issues and impacts that hiring algorithms can have on the hiring process. The report covers the ways in which bias can be baked into hiring algorithms. It’s a refreshing piece of content in a sea of HR tech press that relentlessly praises new hiring technology as efficient and transformative, often ignoring the impact on the candidate.

Honestly I could fill this post with quotes from the report. But I’ll stick to these two bits from the executive summary. If you don’t have the time to read the report, at least read the executive summary. Then carve out 30 minutes to read the rest before 2019. Consider this your last professional development activity of 2018.

Hiring is rarely a single decision point, but rather a cumulative series of small decisions. Predictive technologies can play very different roles throughout the hiring funnel, from determining who sees job advertisements, to estimating an applicant’s performance, to forecasting a candidate’s salary requirements.

Hiring tools that assess, score, and rank jobseekers can overstate marginal or unimportant distinctions between similarly qualified candidates. In particular, rank-ordered lists and numerical scores may influence recruiters more than we realize, and not enough is known about how human recruiters act on predictive tools’ guidance.

After you read the report, go further, and read How to learn about ML/AI as a non tech person.

Could machine learning replace career coaches?

Buried at the bottom of an an HBR post titled 8 Ways Machine Learning is Improving Company Processes, is a little nugget about the ways machine learning might soon affect career planning. Machine learning could help employees in navigate their career development by providing:

Recommendations (that) could help employees choose career paths that lead to high performance, satisfaction, and retention. If a person with an engineering degree wishes to run the division someday, what additional education and work experience should they obtain, and in what order?

Could this be a career coach in the future of work? It’s a fascinating idea and I’d love to see it in practice. We’ve already seen machine learning technology take over some parts of a career advisors job. There’s even a chatbot in development that’s trying to be a career coach (let’s hope they’re better than LinkedIn’s mediocre job recommendation algorithm.) IBM uses AI to guide job seekers through their search.

A good career coach will listen to you, help you work out ideas, guide you through an ambiguous process, support you emotionally, and reflect your own words back to you. Machine learning technology can’t do this yet, in answer to my clickbait title.

But there aren’t enough good career coaches to go around. And few people can even afford a good career coach. Moreover, not every organization offers career coaching that helps employees navigate their next steps. Tools that help people navigate a world full of increasingly ambiguous career paths are mighty helpful.

Like many jobs, career coaches won’t be fully replaced by robots or artificial intelligence anytime soon. There will always be people who prefer working with people over machines. But the role of career coaches will change as new tools and technology emerge. Career coaches need to be aware of these changes. The workplace and available roles are shifting rapidly. Career coaches need to be able to coach their clients through these changes. They need to rethink outdated career advice, especially given that our job search is becoming less human. University career departments in particular need to upskill.

Today’s post is brought to you by my half way mark to 50K words for #NaNoWritMo. I’m deep into a chapter on the future of work for my book and still finding a ton of good content to write about. The challenge of course is to write about it and not just read about it. Reading is not writing, I have to remind myself a bajillion times a day.

If you’re into this type of stuff, subscribe and I’ll send you things about careers, future of work, and probably a bunch of gifs.

The algorithm will manage you now

“Workers increasingly see assignments and wages doled out by artificial systems rather than human managers, and have to rely on AI, not HR, when things go wrong. According to tech experts, the rise of algorithms is changing not only how we earn a living, but who gets access to jobs and other opportunities — if their data checks out — or not.” – Forbes, Algorithms And ‘Uberland’ Are Driving Us Into Technocratic Serfdom

I rarely link to Forbes pieces because their ad game is excessive (even with my ad blocker) but the quote above captures the workplace transformation quite succinctly. From spying on workers, to replacing managers with AI, to using questionable data and AI insights to determine who gets hired, the world of work is changing in ways that need examining fast.

The Forbes article was referencing the book UBERLAND: How Algorithms Are Rewriting The Rules Of Work, which has just rocketed to the top of my reading list. Until then, I’m definitely looking out for the author on the podcast circuit.

 

It was only a matter of time

Students are looking for ways to beat AI recruiting tools like HireVue. And now coaching services are offering help:

“A start-up called Finito claims it can coach candidates to beat AI for as long as it takes them to get a job — but at a total cost of nearly £9,000. Candidates are steered through interview dry runs and get tips on what skills are needed to get past robot selections, in sectors including finance, public relations and the arts. They then watch footage back to spot foibles that could be flagged up as nerves.”

Add helping students beat the AI recruiting process to the list of things career services needs to upskill.

beat AI recruiting

Career Services needs to upskill. Here’s how.

I originally wrote this as a guest post on Switchboard, an alumni platform that connects students and alumni. Switchboard is one of the few ed-tech companies who understand the nuances of higher education transformation. Their higher education innovation fellowship and upcoming conference ListenUpEDU are models for professional development in higher education. And they kill it with good advice for the future of alumni relations

By now we’ve all seen the headlines about the future of work. Beyond headlines about job-stealing robots, the reality is that machine learning and artificial intelligence technology are disrupting career paths. According to the World Economic Forum’s latest report, The Future of Jobs 2018, AI will create 58 million new jobs within the next five years. In a 2017 Deloitte report, Catch the Wave: The 21st Century Career, the authors note that only 19 percent of companies even have traditional career pathways. The future of work is filled with ambiguity and non-linear career paths.

With so much change ahead, career centers need to rethink outdated career training models. Career centers’ primary focus should not be to prepare students for linear careers anymore. Instead, they should prepare students for a lifetime of career changes. Navigating these ambiguous career paths requires students and alumni to embrace upskilling and lifelong learning. This same advice applies to careers services staff too.

Continue reading →

The algorithm will hire you now

AI Hiring

A snapshot of opinions on HireVue on Reddit

It appears the use of AI in the hiring process is finally hitting mainstream awareness. The Wall Street Journal just released a video report about the role of artificial intelligence in the job search. As part of their Moving Upstream series that explores new trends and technologies, the WSJ investigated two companies that use artificial intelligence to decide if you get hired: HireVue and DeepSense.

The video is worth watching, especially if you’re in the job search or working in career services.

The video begins with an introduction to HireVue, a platform that uses machine learning to assess and rank users on their video interview performance. The video provides an overview of the scoring process and the science behind their facial analysis software from HireVue’s chief psychologist. The company uses millions of data points taken from a candidate’s facial expressions, language choice, and tone of voice to measure and determine a candidate’s fit for a job.

There’s a notable part of the video when the journalist asks the psychologist if all interview videos are reviewed by a human. The psychologist chooses his words carefully, noting that recruiters could watch all the videos if they wanted. But we all know that’s not likely. HireVue exists to make the interview process more efficient. Their product is marketed as a way to save time. It’s not efficient if recruiters have to watch every video.

Later in the the video we meet a college student. He estimates that almost half of his interviews have taken place on HireVue. He’s not a huge fan because he thinks it’s hard to show his true self in video interviews.

There’s likely another reason he dislikes it: Interview preparation requires hours of preparation. Thinking on your feet and providing authentic, yet impactful responses, takes a lot of work in the interview process. It’s hard enough knowing you have to impress a human. But knowing a human many never hear your answers is disappointing. It’s the resume black hole on steroids.

The video report includes some welcome skepticism towards new HR tech from Ifeoma Ajunwa, sociologist and law professor at Cornell University. When asked about the validity of microexpressions, she explains:

It’s still a developing science. The important thing is, there is no clear established pattern of what facial expression is needed for any job. Applicants can be eliminated for facial expressions that have nothing to do with the job.”

AI is Changing the Entire Hiring Process

Artificial intelligence isn’t just changing interviews. It’s changing how candidates are hired at every stage of the hiring process. The WSJ video goes on to profile Deepsense, an AI platform that builds a behavioral profile for every person. The company creates a behavioral profile based on social data taken from publicly available data from sites like Twitter and LinkedIn.

The DeepSense AI process

Then they use the data to “run scientifically based tests to surface people’s personality traits.” In a separate article, the cofounder and CEO of Frrole (which developed DeepSense), notes: “One thing people don’t realize is that how little data is required to start making deductions about you, and probably correct enough.”

AI hiring HR Tech

Screenshot of Deepsense dashboard from WSJ video report

Probably correct enough. That’s tough to read when the stakes are so high. The job search is an emotionally exhausting process. Job seekers have families to support, dreams to achieve, health insurance to secure, and bills to pay. They expect to be evaluated fairly and accurately. Probably correct enough isn’t enough in a high stakes situation.

Currently a big five consulting firm is using their service.

The potential for discrimination and bias with new HR technology is high. How do you ensure your public data is correct? How do you challenge the methodology behind the collection/selection of that data? How do you know if you’ve been discriminated against if it’s all done by algorithmic decision?

Beyond the potential for discrimination and bias coded into algorithms, there’s another disturbing bit of information from that video: job seekers may not know they’re being evaluated by an algorithm. As the WSJ reporter notes:

“I go into this knowing something that HireVue acknowledges many job candidates potentially do not. That my responses are being assessed not by human beings, but by AI, analyzing my tone of voice, the clusters of words I use, and my microexpressions.”

Do people know that every post, article, tweet they put on line can now be analyzed and scored as a basis for hiring? These questions, and plenty more, urgently need answers as companies implement new hiring technology.

The quality of your head movements will help determine if you get hired and I’ve got nothing but questions

Yobs.io isn’t the first HR tech company to promise better candidate selection technology through AI and predictive analytics. HireVue has been using algorithms to review and assess video interviews for companies like Unilever and JP Morgan, and they’ve got $93 million in funding to do it. AI technology is rapidly changing the job search.

Yobs.io, however, positions itself as a platform that can identify a candidate’s soft skills and improve team dynamics. Their tech implements “quantiative soft skills analysis in the recruitment.” It claims its platform “determines the emotional state of your candidate which reflect the real-time soft skills that they will take to the job everyday.” Their algorithms analyze facial expressions, word choice and tone, and even head speed to predict candidate success in an organization.

I find it hilarious that employers are banging the drums about the need for employees with soft skills yet they’re increasingly willing to hand over the process of selecting people with those same skills to a machine.

I work on interview chatbots and conversational AI in my contract work. I find it fascinating. I enjoy watching the algorithm improve and seeing its limitations. However, technology that uses personality assessments and predictive analytics to make hiring decisions fills me with questions. They’re questions that I rarely see addressed in tech media or HR industry coverage. They’re questions in need of answers that aren’t marketing copy.

Just look at that engagement level! Source: Yobs.io website

Here’s the ongoing list of questions I never see answers to:

How are companies evaluating whether hires by AI are better than human-led hires? Is this technology trusted for use in all hires, including executive management? Moreover, do the AI engineers have the soft skills they’re designing algorithms for? Does it matter if they don’t? Do the managers who oversee the implementation of this technology also have the soft skills they seek?

Also…

Why should my head speed be part of my interview evaluation? How much weight is my head speed given in the algorithm? What is a quality head speed and how does it affect my ability to do a job that I’ve trained for? Who decides what interview tone is appropriate? Would a monotone AI engineer with an abnormal head speed, a high rate of neuroticism, low rate of extraversion be an acceptable hire (trick question, of course they would, they’re the most in-demand occupation)

And…

Who loses out on an opportunity during the tuning phase of the algorithm? Algorithms don’t work perfectly out of the gate. What feedback loops exist inside the organization’s that use this tech to ensure they’re not getting false negatives? How do HR tech companies who claim to reduce bias prove they actual reduce bias rather than reinforce it?

Humans are flawed. But so are algorithms and even the data we use to build them. Just because it can be measured (head speed) doesn’t mean it needs to be. Asking the hard questions about new technology is important, especially in high stakes situations like job interviews and career progression.

Also, I’m parking this fab find here: Yobs.io uses the big 5 personality traits (OCEAN) to predict candidate fit. There’s a fabulous overview of the Big 5 that includes psych student videos explaining the big 5 concepts. Highly recommend watching these videos, especially when they discuss the person-situation debate.

What makes you trust a chatbot?

Or better yet, should we trust chatbots?

Should we build relationships with AI bots?

Should our children?

Two recent podcasts explored issues of trust and relationships with chatbots and robots.

There’s.so.much.to.say.on.this. I’m writing up a storm elsewhere this week so I’m just parking them here for the curious. You should listen and then get your friends together for a podcast dinner to discuss it. Because it’s a wild topic sure to make for engaging conversation.

Science Friday – A Bot You Can Trust

Radiolab – More or Less Human

Your job search is becoming less human. Here’s how to adapt.

Imagine you’re a job seeker looking for work. You submit your resume to a company’s website.

Your resume is scanned by AI that evaluates your resume against the job description. Then it compares your qualifications to a database of current employees’ qualifications. The algorithm also pulls in some publicly available data about you, like your social media profiles. It scores you based on that data and your resume. Your score puts you above the competition. Your resume isn’t reviewed by a recruiter.

Next you get a text on your phone. It’s the company and they’re asking if you have time to answer a few questions. You answer a few basic questions about your professional experience and interest in the role. You realize it’s a chatbot half way through but you’re just happy to avoid the awkward phone interview.

You make the cut again. You receive an automated email with a link to an online video interview platform and instructions. You record your answer to the interview questions. It’s awkward to stare at yourself on the screen. There are no visual or verbal cues to see how your answers land. Your responses are recorded. An algorithm analyzes the video, reviewing your micro expressions and looking at 25,000 possible data points to evaluate your personality and fit within the company. Your video response is scored by the algorithm.

Then you get an email from the recruiter. You’ve passed all the steps. They’d like to invite your for a day in the life experience at their company.

The visit is the first and last opportunity you’ll have to interact with a person in your entire job search.

Back to reality. The scenario above isn’t totally hypothetical. It’s reflective of the current hiring process evolution. Companies are increasingly adopting HR tech that uses AI to automate the hiring process and make it more efficient. For example, here’s what hiring looks like at Unilever:

Candidates learn about the jobs online through outlets like Facebook or LinkedIn and submit their LinkedIn profiles — no résumé required. They then spend about 20 minutes playing 12 neuroscience-based games on the Pymetrics platform. If their results match the required profile of a certain position, they move on to an interview via HireVue, where they record responses to preset interview questions. The technology analyzes things like keywords, intonation, and body language, and makes notes on them for the hiring manager. All of this can be completed on a smartphone or tablet.

If the candidate makes it through these two steps, they are invited to a Unilever office to go through a day-in-the-life scenario. By the end of the day, a manager will decide whether they are the right fit for the job.

A fundamental shift in hiring is under way and it’s powered by machine learning. From resume screening by AI to interview chatbots to predictive analytics that determine who’s most likely to leave a job, the list of startups transforming the hiring process is long. Over half of HR tech investments in 2017 went to companies offering products and services powered by AI. Companies like Entelo, an AI recruiting platform, use machine learning to determine whether you’re a fit for an organization. Entelo’s knowledge base provides a few hints on how the AI will evaluate you:

The shift to automation is making the hiring process less human. As a job seeker it’s not always obvious when AI is used as part of the hiring process. You might not know if your professional qualifications are being evaluated by a human or an algorithm. To stay competitive as the hiring process evolves job seekers need to stay informed and adapt as new HR technology enters the market.

Here’s how to start.

Get curious about HR Tech

Explore the range of new HR technology that’s being used in the hiring process. Get curious about how these tools are used. Then experiment with new HR technology that also helps job seekers. Tools like Jobscan and VMOCK are valuable resources that use machine learning to help your improve your resume. There’s even a promise of a chatbot to help you navigate your career.

Next, research which companies are using machine learning for hiring so you can prepare accordingly. Right now big companies with large resume volumes are the ideal automation customers. Smaller businesses and startups aren’t using them as much yet. Some HR tech products list which companies use their services. Before you apply to a job, email a recruiter or ask a current employee about their hiring process so you know up front whether you’ll be engaging with a machine or a human.

600+ companies in 140 countries use HireVue.

Be prepared to go beyond resumes

The resume isn’t going away any time soon but the application process is evolving to evaluate you on more than your resume. Instead of submitting a resume, candidates are taking part in hiring assessments like Pymetrics, a collection of that neuroscience games that “collect millions of data points, objectively measuring cognitive and personality traits.” Tools like Entelo assess your social media data as part of the application process:

AI Recruiting on Entelo

Creating professional content so the HR bots can find and evaluate you could make you a more competitive candidate than a resume alone. Start by producing small bits of content online. Create a personal website, show off a portfolio online, write short blog posts, or share articles on Twitter related to your professional interests to be seen by the bots.

Ask hard questions about AI and HR technology 

There are plenty of ethical questions we need to ask about AI and reinforcing bias in recruiting. Job seekers can contribute by asking hard questions too. Sometimes it’s as simple as asking how.

How do algorithms score candidates? How are candidates screened out of the process? How do candidates rank if they don’t have online profiles or publicly available data for algorithms to find? How would a candidate beat the AI system? How much do hiring managers trust their AI recommendations and scoring? How do these platforms reinforce existing bias?

Then ask yourself the hard questions: Are you getting all the information you need in the hiring process – company culture, opportunity for growth, management styles – to make an informed decision? Does an automated candidate experience make you more or less likely to want to work for a new company?

Become an actor

One question they get frequently, said Lindsey Zuloaga, director of data science at HireVue, is if an applicant is able to trick the A.I. Her answer: “If you can game being excited about and interested in the job, yes, you could game that with a person as well,” she said. “You’re not going to game it without being a very good actor.”

Employers seek candidates with strong soft skills. As more employers delegate emotional intelligence screening to automated tools you need to ensure you’re expressing that emotional intelligence. Start by recording yourself so you know how you look, talk, and express yourself on screen. Pay attention to your tone, body language, and facial expressions. Learn how to build your soft skills to improve your emotional intelligence. Spend more time interacting with people to improve your communication skills outside of digital environments. You might even want to take some acting or improv lessons to get comfortable showing those necessary emotions.

Cultivate those professional relationships

Will recruiters eschew a recommendation from a human in favor of their AI scoring system? Do AI hiring platforms incorporate internal recommendations into their scoring model? We don’t know. So for now we can assume that internal referrals via professional relationships might be a way to beat the algorithms (or at least, get around it). More importantly those professional relationships take on greater importance the more automated the hiring process becomes. Conversations with people inside of companies give you valuable insights. Discussions with current employers also give you a feel for company culture and management style, making up for the insights you lose in an automated process.

Sharpen your persuasion skills 

We’re not in a fully automated hiring process (yet). Job seekers still have a chance to engage with humans during their search. But the hiring process is evolving and making some career advice outdated. When you finally get in front of an employer it might not be what you expected (i.e. those behavioral interview questions you memorized might not be as relevant in the future). But one thing won’t change: once you engage with a human you still have to persuade them that you’re the best person for the job. Your job search has always been an act of persuasion. That much hasn’t changed. After you learn the new automated systems focus on building your persuasion skills. Reflect on what the companies needs and how you meet that need. Learn how to tell an engaging professional story that connects your interests to your future team’s needs. Show employers your intellectual curiosity and passion as you ask questions about the role. Seek out new conversational opportunities so you get better at engaging with people from different backgrounds.

We all need to pay attention to the way hiring is changing. With millennials looking at a lifetime of job hopping, we’re going to have adapt fast to new hiring processes. The traditional way of doing things won’t always work. As this article so cleverly points out:

“those first impressions so carefully emphasized by career coaches are now being outsourced to artificial intelligence.”

How much should this AI Chatbot Writer job pay?

Hybrid jobs are all the rage currently and are some of the top paying jobs in the market right now. If you’ve got soft skills, business acumen, and technical skills, you’ve got the ticket to a high paying job.

Hybrid roles are super interesting to follow because they are so new. Their descriptions and responsibilities differ from one organization to another. This is particularly the case with AI interaction designers, a emerging job category I’m paying a lot of attention to lately (in part because I’m slightly obsessed with chatbot design as of late.) Diane Kim, who designs the friendly virtual assistant bot at x.ai, summed up this emerging field in her interview with Wendy and Wade, a career advising chatbot:

“The fact that AI Interaction Design is so new gives me the freedom to be experimental. I also have the unique opportunity to be part of defining an entirely new field. This is actually both what is most exciting and most challenging about my job…But it’s challenging because none of us really know what this is yet — we’re all figuring it out together. It’s really different from, say, being a recent grad in your typical UX role for a visual interface, with decades of research and best practices to follow. We don’t have the same industry standards or guidelines yet for conversational design, but the fun part is figuring them out as we go.”

So it’s within that context that I examined this AI chatbot writer role from JustAnswers.

Chatbotjob Chatbotjob

The skill requirements on this role are massive. Let’s break it down.

  • You need quantitiatve and qualitative skills
  • You need to be a seriously good at writing (perfect tone!)
  • You need to understand Sales (identify (and contribute to?) revenue opps!)
  • You need be an experimenter – test and retest
  • You need mad research skills
  • You need the collaboration skills to work with diverse teams
  • You need to understand user experience
  • You need to dive into professional fields that requires years AND be required to anticipate which quesitons users would ask AND write the answers.

This is one hell of a robust skill set. That last ask – expert with diving into deep professional fields like medicine and law – really threw me off. Who is this person? And will you pay them a shit ton of money for this expertise and skill set?

It’s likely this job is like most job postings: crammed with all the ideal things. There is probably flexibility – an applicant doesn’t have to have all those things.

I’m curious about how much this role pays because writing is an underpaid profession. Some managers who don’t write assume it’s easy – after all they write emails and reports! Copy is everywhere and people assume it’s easy to produce. Thoughtful copy – the kind that strikes the perfect tone! – takes time and creativity to produce. People in quantitative fields tend to overlook that.

But bad writing, especially in AI conversation design, leads to awkward interactions with the product. For example this was my recent convo with a new recruiting bot Robo Recruiter:

If writing is underpaid but AI is a hot hot hot field, how much should we be paying our AI chatbot writers?

I’m crowdsourcing your answers below in the comments: how much do you think this job pays? Do you think it pays as much as a machine learning engineer? As a product manager?

Write your answer below.

Then see what Paysa pegs the going salary rate in San Francisco.