Building Diversity in AI: 3 New Hires Discuss Supply and Demand in the Tech Revolution

For all the worry about AI replacing human workers, what actually seems to be happening is a surge in hiring for high-paying AI-related jobs.

Thousands of enterprises across all types of industries need AI workers who can build the algorithms to produce text and images when prompted, or train AI models, to name a few highly desirable skills. Still, it’s a tall order to find candidates with not only the right skills but also the diverse backgrounds to minimize biases that can creep into AI systems.

Both are required, experts say.

The future of IT and security is autonomous. But most organizations don’t know which manual processes are easy to eliminate. This is where you start.

“If you don’t include a wide swath of human beings in the creation of your technology products, when you fail — because it’s not an ‘if’ — you will lose money because you’ve spent all of this money on development without considering the human beings at the end,” said Karim Ginena, founder of AI governance and research consultancy RAI Audit, in a Knowledge at Wharton article last year.

If you don’t include a wide swath of human beings in the creation of your technology products…, you will lose money.

Cybersecurity leaders have long known that diverse teams keep us safer. It turns out they also make our AI smarter – and boost the bottom line, experts say. So how do organizations find these essential new workers?

It may be worth following the feds. The goal of creating more effective and ethical AI systems is one reason why the federal government is actively recruiting a diverse AI workforce in terms of gender, race, and ethnicity. The Biden administration plans to add more than 500 AI professionals to the federal civilian workforce by the end of fiscal year 2025, with the hope that many of those new hires will come from demographic groups not traditionally well represented in STEM. And they’re not just hoping. The administration is highlighting the difference between AI and “AI-enabling” jobs (data scientists, analysts, and other roles that are less technical), thereby broadening their pool of candidates, and participating in diversity-centric job fairs and gatherings like October’s Grace Hopper Celebration, a conference for women and non-binary workers in tech. Congress, too, is doing its part, looking to expand more AI research and development funding to institutions like historically Black colleges and universities.

Focal Point talked to a collection of new AI hires – individuals from diverse ethnic, academic, and professional backgrounds, all who are relatively new to the field of AI – to hear about their experiences getting noticed, getting hired and onboarded, and adjusting in the workplace. Their stories can help inform enterprise leaders and those building teams for the new tech revolution. The smoother we can make the employee experience for AI workers, the better – after all, these are the folks on the front lines of AI, who are experiencing the cultural shift that all enterprises will have to make to implement effective AI adoption, direction, and governance.

Diversity in AI starts by getting in the room

Lacey Miller, head of growth marketing at Perigon, worked in marketing for a number of startups over the past decade because she loved being a “foundation part of building a business.”

In her last job, at a 3D design company, Miller (at left) had the chance to be part of building something new – in that case, incorporating generative AI in a game-changing way to help artists and other creatives enhance their designs. She was instrumental in integrating generative AI into a tool that uses text prompts to find images that are dragged into background scenes. As soon as that skillset hit her resume, recruiters came out of the woodwork to talk to her. She knew that she wanted to be building the AI tools, not just using them. Her job at Perigon, an AI-powered media aggregator located in Austin, Texas, involves setting up AI for marketing tasks and to identify sales targets most likely to need the technology that uses real-time media.

Even though she was highly sought-after for other positions in marketing, she was excited by the challenges this particular opportunity offered, so she went after it. She reached out to the co-founder of Perigon and sold her skills. Because machine learning is so new and generative AI is such a vulnerable yet vital technology, Miller says it’s important to show potential employers that you can learn the skills needed to utilize it.

I want to hear what they’re talking about and learn to speak their language so when I go out to talk to customers, I’m informed.

Miller wants to learn more, but one of her biggest workforce challenges is to find her place in the room with male developers. It’s a different culture from her marketing background, and she’s found that she has to invite herself into meetings about building the technology. “I want to hear what they’re talking about and learn to speak their language so when I go out to talk to customers, I’m informed.”

[Read also: Preventing AI bias starts at the top – just ask these female chief AI officers

She also sees her role as an opportunity to address some of the biases found in current AI models – for example, the prominence of sweet, female-sounding voices in chatbots. Miller recognizes the strategy behind it, and how it creates the perception of an attractive woman selling the product. It is a skewed perception, for sure, in an already male-centric profession, and something she hopes she can raise awareness about as she progresses in the field.

Miller straddles a so-called soft-skill job (marketing) with a hard-skill role (building AI platforms). Her willingness to learn the technology skills to work in AI has opened doors for her, but she knows that others following a similar path from a non-tech job to AI may find they get knocked down along the way. She encourages them to persevere and have faith that they’ll eventually find the place where their unique voice in the room will be heard.

Timing is everything – especially when you’re early

Sujan Abraham, senior software engineer with Labelbox, has spent most of his career so far working with search engines, building the systems to make searches happen successfully. Labelbox, a San Francisco-based data factory for generative AI workloads, needed someone with experience working on search systems.

“Labelbox has a system where they bring in a lot of data, including millions of images and videos, which need to be searchable,” says Abraham. His expertise matched Labelbox’s need to build a search platform, and he notes it was a straightforward hiring process.

In his job, Abraham uses AI to build the search engine. While he had a basic understanding of the technology, he’d never used it in his work before. Now that he’s got experience building AI systems into search programs, Abraham says that, like Miller, recruiters have been reaching out to him.

Adjusting to the rapid pace of technology, and the emotional rollercoaster of AI, has been the real challenge for Abraham, whereas fitting into his role in the company has been smooth. Coming into a small organization with a defined skill set “helps in eliminating the cultural issues that can creep in,” he says.

The way AI is used now is pretty generic, but in the next few years, that’s going to change, and the demand for people with AI skills is going to increase.

Abraham feels fortunate he was in the right place at the right time, not only for his career move but also to be working with generative AI in its infancy.

[Read also: AI isn’t likely to replace your job – at least not if you work in SecOps]

“The way AI is used now is pretty generic,” he says, “but in the next few years, that’s going to change, and the demand for people with AI skills is going to increase.”

Mentorship matters

Zanele Munyikwa, economist at Revelio Labs, recently earned a Ph.D. in economics from MIT, with her field of study crossing the intersection of economics and computer science, focusing on economic impacts from the labor perspective of different AI technologies.

In her current role, she uses AI in several different ways. As an economist, she helps data scientists come up with different measures for data models. “One model we worked on was [a] ‘likelihood of hiring’ model, which is looking at job postings and measuring the predicted number of hires that would result in that job posting,” explained Munyikwa. “For my role, it is applying genAI for specific use cases to generate reports.”

What’s been most valuable for me is building a support network of peers and people further along in their career.

While working on her Ph.D., Munyikwa also spent time networking, building relationships, and finding mentors. She participated in a data boot camp with Correlation One, designed to prepare diverse talent for data and analytics and AI roles. She was mentored by people already working in AI or data science fields. She also attended career fairs and conferences to prepare for her job search.

[Read also: 3 eye-opening documentaries on AI to watch this weekend]

Munyikwa’s hiring process into her current job with Revelio Labs, a workforce analytics firm based in New York, was complicated and involved several layers of interviews. Her boot camp and mentorship prepared her for the experience.

“What’s been most valuable for me is building a support network of peers and people further along in their career,” said Munyikwa. “That network can be helpful when you end up in environments that aren’t particularly welcoming.”