Expert_opinion

Husayn Kassai: Securing Customers Identity with Machine Learning

Husayn Kassai: Securing Customers Identity with Machine Learning

Tackling security-oriented issues has always been difficult for organizations to deal with. Though digitization has almost taken over most sectors, identification recognition systems are still frozen in an offline world. Husayn Kassai, CEO and Co-Founder at Onfido, has been implicating machine learning to deliver next-generation identity verification.

Below are the highlights of an interview conducted between CIO Look and Husayn:

Give a brief overview of your role in the company and how do you diversify your tech solutions that appeal to your target audience?

Onfido’s machine learning technology verifies an ID document, and uses facial recognition to compare a selfie-photo or video with the photo on the ID, be it a passport, a driving license or a national ID card. The identity can then be cross-referenced against international credit and watch-list databases.

The company supplies several identity verification solutions. These include our Document Check, which ensures that users’ identity documents have not been forged, digitally tampered with, lost or stolen, and Facial Check, which compares the photo on a user’s identity document with a selfie to make sure users are who they claim to be. The technology enables organizations to perform checks on over 6000 document types detecting anomalies automatically, and using human experts to verify outliers.

What were the past experiences, achievements or lessons that shaped your journey?

I had the idea for the business since my parents moved from Iran to the UK, back when I was 10. It took them several months to be able to rent accommodation in their own name or open a bank account because as new migrants, they weren’t on any credit reference agencies. At that point I thought that credit bureaus must be magical places – but I later realized they’re just databases, and due to hacks and breaches, not very secure ones. It became clear that a better solution was needed for people to be able to verify their identities and gain access to these crucial services, and so Onfido was born.

What were some of the primal challenges and roadblocks that you faced during the initial phase of your journey?

Onfido has grown rapidly over the last six years, and that brings with it a number of challenges. Expanding into new international markets means navigating new and complicated regulations. It also means working across different time zones, which makes communication and teamwork difficult. Thankfully, we have a very strong culture and a world-class team, which makes it easier every time we enter a new market.

Kindly take us through your vision or approach towards business.

I have a fundamental view/belief that pure talent (natural aptitude or skill) is randomly distributed in the population and that the essence of entrepreneurship is about (a) recognizing true talent (not CV scanning for grades or tick-box norms) and (b) facilitating a space for the talent to flourish (culture). Despite the “tech” buzz-word, it is all about the people, and to attract talent, entrepreneurs need to overcome misconceptions or prejudice/biases.

With ascending in the number of pivotal and groundbreaking technological advancements the role of technologists is continually evolving. How according to you, has this role changed over the years?

I didn’t start out as a technologist. We had the idea for the business based on our own experiences, and tech was clearly the best way to solve the problem. Onfido uses Machine Learning technology, which grew out my technical co-founder Ruhul Amin’s Computer Vision and Machine Learning thesis.

Tech is a challenging sector as so many young startups fail. But it’s also an incredibly exciting industry to be a part of – especially in London, which is unquestionably the FinTech capital of the world. There’s a huge network of innovative companies working together here, and fantastic collaboration between regulators, investors, businesses and talent.

Describe some of the vital attributes that every tech individual should possess.

Learn to recognize and hire true talent. Don’t ‘CV scan’ for grades or tick-box norms, but look for natural aptitude or skill. Having the right people onboard is the only way to ensure success. Work collaboratively. Welcoming diverse viewpoints helps come up with the best decisions and deliver on them.

Where do you see yourself in the near future and what are its future goals?

Our vision is to create an open world where identity is the key to access. The world is moving online, but the means to carry out identity verification is frozen in time in an offline world. Millions of people who don’t have a credit history – or just have a thin one – are unable to access countless everyday services. With our identity verification technology, those underserved people can access online services they need, and businesses can gain access to a whole new group of trusted customers. As identity verification technology continues to evolve, it will become the key to remotely unlocking a huge range of global goods and services, from banking and travel services through to voting and healthcare.

What is your advice for the emerging tech enthusiasts?

Don’t settle for the status quo. We know that current processes for identity verification simply aren’t good enough. Thanks to new tech developments, we’re able to change that – but only through constant curiosity, research and iteration.