Listen to this story
“How big is the universe?” asks Alicia Nash as her face beamed with curiosity and allure. “Infinite. I know because all the data indicates it’s infinite,” answers John Forbes Nash Jr. with confidence even though there is no evidence to support his statement. “I don’t; I just believe it,” he says with a rather innocent smile.
Though Ron Howard’s A Beautiful Mind focused loosely on Nobel prize winner John Forbes Nash’s battle with schizophrenia, it did point to his unique ability to see patterns where no patterns exist. He viewed the world in a different light, and that was all he needed to make his mark in history.
Sign up for your weekly dose of what's up in emerging technology.
With nearly 15 years of research, development and management experience in data science and software development, NICE Actimize’s Chief Data Scientist Danny Butvinik. NICE Actimize is a software company that helps its customers in combating financial crimes.
At Actimize, Danny builds and manages applied research-based ethical AI within Actimize FinCrime portfolio products and its supporting services, leveraging collective intelligence and data consortium to provide the clients with transparent, adaptive, and measurable analytical solutions by reducing fraud losses, time-to-insights and by leading the way for Actimize clients and market. He takes a central role in the processes of due diligence, IP development, engagements with clients, business strategy, and scientific research roadmaps to deliver short- and long-term goals in a fluid environment. His experience in the field has also earned him the reputation of being a mentor, nurturing some of the brightest minds in data science and AI.
With 12 pending patents, he aims to define and implement world-class data science practices to ensure those insights are timely, robust, repeatable, and trustworthy.
In an exclusive interview with Analytics India Magazine, Danny walks us through his professional journey from academia to industry.
AIM: What drew you to data science?
Danny Butvinik: My perception of data science is quite perplexing. We have digital devices all over the place: the internet, emails, social platforms, houses, streets, transportation, aviation, satellites, mobiles, watches, etc. We are literally saturated with abundant digital organisms that live their lives and leave their footprints. In fact, this is not their footprints; it is our footprints. These devices trace, store, capture, send signals, track, detect, and identify whatever they intended for. Every day we send 306 B emails and 500 hundreds million Tweets. In 2020, humanity generated 2.5 quintillion data bytes daily. By 2025, we will generate 463 exabytes of data each day. If we take a step aside and look at our world and the enormous amount of information passing through the veins of the digital world, we perhaps may see or hear some noise, a lot of noise. That said, there are shapes, patterns and trends in this presumed chaos. You just only need to know how to look at it. And once you do, you will reveal an unimaginable structured system that can provide insights you never thought possible.
But this is only the beginning. Once you know how to make sense of the infinitely complex data systems, you can go one step further: predict. Once you are capable of prediction, the next step would be prescription. Prescription is about being able to ask and answer the question: “What needs to be done so that we get a desirable result.” All these aspects drove my interest in data science.
As mathematics is a language through which we can describe nature, data science is a field through which we can understand our intricate, perplexed and mysterious world of cause and effect, relationships, trends, and patterns.
AIM: What was your first job in this field? What were the key takeaways?
Danny Butvinik: My first work was in academia, where I dealt with multidisciplinary research. My key takeaways were that data science is like a newborn, taking baby steps, despite the fact that all of its pillars stand on the shoulders of the giants (mathematics, statistics, information theory, computer sciences, theory of neural networks, the computational learning theory and other.)
AIM: How did you pivot to an industry role?
Danny Butvinik: My long journey to Chief Data Scientist originated during my years in the academy, where I researched advanced statistics, information theory, computational geometry, data structures, streaming algorithms, advanced simulations and optimisations. My solid mathematical background allowed me to explore various fields and draw precious experiences that later shaped my competence in Artificial Intelligence and Data Science.
At some point, I decided to exploit my deep theoretical knowledge by materialising it in the industry. Before I joined NICE Actimize, I’d been with several companies in different domains such as computer vision, image processing, signal processing, security and healthcare. On top of my strong mathematical background, I developed my knowledge in various scientific disciplines and cross-domain industries. This shaped my perception of data science as a discipline and paved the way for things I would love to do in the future. I delved into Incremental Online Machine Learning, Online Active Machine Learning, Online Reinforcement Machine Learning and complex AI-based systems.
Having discovered my specialisation within AI and Data Science, I continued to explore and delve into the ‘esoteric’ sub-fields such as Chaos expansion in complex systems, AI uncertainty, parsimonious models, ergodic processes, causal inference, and information-based uncertainty in decision boundary for classification problems.
After I joined NICE Actimize, I realised very soon that financial crime is the most challenging and fascinating domain I’ve been through, and it provides a huge potential for exploration for data scientists. It resonates perfectly with my favourite research topics as well.
My main drives are curiosity, enthusiasm and a voracious desire to quench the thirst for knowledge. Of course, that means a lot of hard work, reading, continuous learning and resilience.
AIM: Tell us about Actimize and what it offers.
Danny Butvinik: NICE Actimize is the leading worldwide provider of financial crime, risk, and compliance solutions. Actimize has different LOBs, including AML, Fraud & Authentication Management, Financial Markets Compliance, Investigation & Case Management and Data Intelligence.
We leverage machine learning and AI to detect and prevent financial crimes across the financial services industry, including some of the largest global financial institutions.
We exploit elements of decentralised AI such as federated machine learning to leverage data consortium in fraud detection over the cloud and create a paradigm shift in stopping real-time fraud before it happens by exploring online incremental machine learning and obtaining continuously adaptive solutions for financial institutions.
AIM: As the Chief Data Scientist at NICE Actimize, what are your responsibilities?
Danny Butvinik: I serve as the principal professional data science authority for the organisation. I lead the company’s efforts in advanced analytics and autonomous financial crime and compliance. In addition, I lead the Actimize community of Data Science to practice creating channels of collaboration, knowledge sharing, mentoring and continued growth of the data science practice and practitioners. My team and I also work with the marketing and sales guys to offer insights on customers’ needs and demands.
AIM: Which is/was the most professionally challenging point in your career?
Danny Butvinik: Most challenging and, at the same time, most interesting time in my professional career is now. I’m working with my team on the research of online incremental machine learning for fraud detection. To me, it’s quite appealing to combine such a theoretical approach with real implementation and embed it into production. Of course, being an enthusiast and passionate about what I’m doing takes me through tough moments. But, at the end of the day, I believe in what I’m creating, and that matters.
AIM: What’s your personal goal as a data scientist? What do you want to achieve?
Danny Butvinik: My short-term goals are to establish a solid theory around Online Incremental Machine Learning under certain constraints and bring it to realisation.
My long-term goal is to write a book that combines the Financial Crime domain and advanced science. I see this book to be intended for a wide audience. It will contain different layers for diverse readers.