AI training program helps close gender gap, bringing fresh faces to Europe’s digital workforce

When she entered a technology university in Valence, France to study computer science, Audrey Roumieux was in a first-year group of 84 students. Six of them were women.

“We had no problem finding each other,” she recalls with a laugh. “In the class of 84 people I was one of the first people to be spotted.”

She says that aside from an occasional tasteless joke from her male classmates, she thrived at the university. “I don’t see any real reason why there are few women in IT because I am just as capable as a man to do my job.”

Roumieux went on to another year of university in Marseille and then to an internship at a cancer research center that was developing machine learning models to detect cancer cells. She became fascinated by the technology; it was statistically better at the task than the doctors.

“It’s almost like magic, and I wanted so much to know what’s behind it,” she says. “I wanted to continue my studies, and that’s when I discovered the Microsoft AI School by Simplon.”

Roumieux was in the first class of the school, which opened in 2018 to help women, refugees, people with disabilities and those seeking second careers enter the world of AI. She now works as a data engineer for Avanade, a company formed by Microsoft and Accenture in a joint venture in 2000.

The Microsoft AI School by Simplon is part of Microsoft’s efforts to bring more diversity to the digital workforce and to fill significant labor and gender gaps. A 2021 report by the European Commission estimates that Europe will need 20 million information and communications technology (ICT) specialists by 2030. Right now, there are 8.4 million ICT specialists working in the European Union, and 81.5 percent of them are men.

In this effort, Microsoft has partnered with Simplon, a social enterprise that specializes in digital training for jobseekers from diverse backgrounds. Simplon has schools in France, Belgium, Ivory Coast, Gabon, Morocco, Senegal, Tunisia, Jordan and India. The company’s schools have trained more than 25,000 people, about 40 percent of them women and about 44 percent without full or any college degrees.

France’s national unemployment agency is the main conduit for new trainees in the AI program (and Simplon’s other digital training programs in France as well). Jobseekers with interest and potential are encouraged to enter the program, and the French government pays most of the tuition.

Louise Joly is the administrator of the AI program at Simplon, which is celebrating its five-year anniversary in March. There are now 20 of these AI schools in France, which have trained about 900 people, about 33 percent of them women. Joly says that Simplon has long wanted to reach gender parity in its classes, but that a variety of barriers holding back women have made it hard to reach that goal.

“At Simplon, we’re less concerned about the reasons behind that glass ceiling than finding a solution,” she says. To that end, it created prequalification course for women only in the AI program, which she says has resulted in a higher completion rate.

She also cites the schools’ hands-on methodology for its success in placing its students into good jobs. Participants in the AI training spend four months in the classroom, seven hours per day. Then they spend a year in an apprenticeship at a company, with one week in class and three weeks on the job. Simplon works with companies to create and continuously improve the training program. Companies select candidates for apprenticeships from Simplon from the beginning of the training program. In this way, Simplon supports companies in a non-traditional recruitment process – producing strong recruits outside of the normal channels, such as level of education, professional background and previous experience.

Laurent Cetinsoy is a professor in the Microsoft AI School by Simplon in Paris, and he is an advocate of the school’s hands-on approach.

“The idea is, if you want to learn tennis, the best way is not by hearing some old guy speaking about tennis for three hours,” he says. “We try to put the student into action as soon as possible, but that doesn’t mean I don’t explain things.”

Even during the intense class portion of the course, Cetinsoy says, the participants work on real projects.

In the first year of the program, he says, the class helped an inventor improve a machine that recycles plastic immediately to be used in an attached 3D printer to create new objects. Class members used AI in training the machine to recognize and sort the plastic by type. “We had the luck to see [Microsoft Chairman and CEO] Satya Nadella visit the school at this time, and he really liked the project,” he says.

Stan Briand graduated from the AI School in February of 2022 after a year spent in an apprenticeship at LACROIX in Rennes. There, he developed a system using AI and the Internet of Things (IoT) to help detect leaks in the water system of the town of Nevers.

“On average in France pipelines are leaking 22 to 25 percent, so about one fourth of the clean water we produce for drinking goes back into nature,” Briand says. “This is due to multiple factors, but the most important one is the pipelines are aging. Aside from the environmental waste, this loss costs huge amounts of money.”

The system Briand developed takes data from 200 sensors that detect water flow in the Nevers water network and uses an AI algorithm to analyze it. The AI helps pinpoint where leaks are most likely to be found. Scrutinizing that data was taking up well over an hour a day for a worker at the water utility. Briand says the job now takes five minutes and frees up time for other critical maintenance tasks.

“France and some other countries are having problems with drought,” making water resources more precious, Briand says. “This is a step toward solving those kinds of issues, so it’s really rewarding.” The system he created is now in use at a second municipal water system and will be deployed to others.

Briand’s supervisor at LACROIX is Reynholds Reinette. During Briand’s one-year apprenticeship, he showed skill and initiative in developing the water anomaly detection system, Reinette says, and it was an easy decision to hire him.

“The project was great, and we wanted to continue working on it,” Reinette says. He says LACROIX recently hired another Microsoft AI School by Simplon graduate in Rennes and will have another apprentice from the school starting later in the spring.

Before training as an AI specialist, Briand taught English in Chengdu, China, and eventually became an administrator at the national level for a group of English-language schools. He computerized the administration system and became fascinated with what he could do with data. When he moved back to France, he decided to seek out training to enter the world of computer programming.

“At that time, I felt that the AI field would require a lot of experience, a lot of math background and basically I thought it would be too hard for me to get into that,” he recalls. “But the Microsoft-Simplon school doesn’t require that much math or even a master level in technology. So, I applied, I passed a few tests, and I got in.”

Cetinsoy, the AI professor, echoes Briand, saying that the most important factors for success in the program are motivation and willingness to work very hard.

“You need an analytical mind and an analytical view to solve problems,” he says. “Good programming skills are more important than being a crack at math.”

Microsoft and Simplon have collaborated on another training program that just began in November of 2022: A specialization in cybersecurity, another area where more capable workers are urgently needed in Europe and elsewhere. In the past three months, high-profile cyberattacks on telecommunications networks in Portugal, government servers in Poland and oil facilities in Germany, Belgium and the Netherlands highlight the problem.

The Cybersecurity School is at the heart of Microsoft’s Cybersecurity Skills Plan in France, which aims to support the training of 10,000 new cybersecurity professionals by the end of 2025.

Sixteen students in the first class of the program have completed their coursework and are beginning 16-month apprenticeships at seven different companies in France.

When Roumieux finished her period of intensive coursework at the Microsoft AI School by Simplon, she began an apprenticeship at Azeo, which turned into a full-time job. Azeo was later bought by Avanade.

She says she likes many things about her job, among them that her clients and their needs are varied; she’s always learning something new. She’s worked on creating ways to analyze and visualize sales data for a large wine and spirits corporation. For an advertising firm, she helped create a program to automatically send emails to clients who chronically make late payments. She’s currently working on organizing and visualizing data for a company that runs company cafeterias and distributes meal cards.

“For me technology is a way to help people in their lives and their businesses,” she says. That belief comes from personal experience, she says.

“I am dyslexic, and when I was little, I was given a little computer to help me with spelling and grammar,” she says. She saw that computer as an ally and realized its potential for solving all kinds of problems.

“Clearly, when someone gave me that little computer, it helped me a lot, and I told myself that’s a tool that helps people, and why not see if I could work with it to do the same for others.”

Top image: Audrey Roumieux, data engineer at Avanade in Paris and one of the first graduates of the Microsoft AI School by Simplon. Photo by Chris Welsch for Microsoft.

Read this story in French here: Un programme de formation à l’IA qui encourage la diversité dans le secteur du numérique – News Centre (microsoft.com)