The Human Element In Data Economy: Dinis Guarda Interviews Prayson Wilfred Daniel, Director Of Transformation Lab At NTT Data Solutions

Dinis Guarda interviews Prayson Wilfred Daniel, Principal Data Scientist and Director of Transformation Lab at NTT DATA Business Solutions Nordics, in his YouTube Podcast. In this episode, they discuss various aspects of data science, digital transformation, and the intersection of technology with philosophy. The podcast is powered by Businessabc.net and citiesabc.com. 

Prayson Wilfred Daniel is the Director of the Transformation Lab at NTT DATA Business Solutions Nordics, where he has been working since December 2023. With over a decade of experience in data science, machine learning, and artificial intelligence, Prayson focuses on using machine learning technologies to drive business transformation, boost customer and employee loyalty, and improve operational efficiency.

He leads and mentors teams in Data Science, MLOps, and DevOps, helping integrate advanced technologies such as Natural Language Processing (NLP), Computer Vision, and Bayesian decision-making into various business applications, including healthcare, insurance, financial services, and digital commerce. His expertise spans a wide range of programming languages and tools like Python, Rust, C++, Terraform, Docker, and Kubernetes, and he is deeply involved in developing intelligent, data-driven business solutions.

During the interview, Prayson Wilfred Daniel told Dinis Guarda about the transformative power of combining data with human insight: 

"Data can be shaped to tell any story, but it’s the people who work with it that truly add value. Data on its own is just numbers; it’s when combined with a brilliant mind that it transforms into something meaningful and powerful."

The essence of human-centricity in data and algorithms

Prayson Daniel emphasises the importance of starting with people and their goals before using data and algorithms to solve problems. It's about leveraging technology to enhance human capabilities, not just focusing on data for its own sake. He tells Dinis:

“My primary focus when working on projects is always people first. Whenever we develop algorithms or perform A/B testing, I start by asking, "What are we trying to achieve?" I begin with the client's goals, not the data. Once we understand the objective, we then ask if we have the data to support it. Ultimately, it's not about the algorithms or the data themselves—it’s about how we can use these tools to enhance our ability to solve problems. Whether the goal is improving quality of life or increasing profits, the key is to leverage what we have to elevate who we are. We don’t build just for the sake of building. It’s about transformation. Define the mission first—where are we headed? Once that’s clear, data and algorithms become the means to help accomplish that mission.”

Understanding data generation for ethical modelling

Prayson Daniel emphasises the importance of understanding how data is generated to identify and mitigate biases in models. During the interview, he highlights it to Dinis:

“When applying machine learning or algorithms, it’s essential to take a step back and understand the mechanism behind the data generation. In Bayesian modelling, we start by defining our prior assumptions and then update them with new data to produce accurate outcomes. This approach helps us examine the processes that generate data, which is critical in identifying potential biases and errors. For instance, if we know a model’s data comes from a specific geopolitical region, we can account for its inherent assumptions and biases. By focusing on the people or systems that produce the data, we can better mitigate pitfalls in modelling, avoiding biased outcomes or exclusion. Ultimately, while data is often referred to as the "new oil," the real value lies in understanding the mechanisms behind it. That’s where the magic happens—allowing us to avoid classic mistakes and build models that are fair and accurate.”

Collaborative governance with IBM watsonX

Daniel shed light on the complexities of managing multinational layers of data and navigating the nuances and strategies inherent in such a task. The discussion delved into the pivotal role of IBM watsonx in this landscape, highlighting its unique ability to facilitate governance integration and ethical considerations, crucial for protecting companies and ensuring regulatory compliance.

He said, “I think the biggest part that we've been working with IBM is on the governance so as you mentioned from coming from NTI, we work with different kinds of clients at different sizes and they are in different kinds of technological journeys. So we have models deployed everywhere and this led us to uh have to make a partnership with IBM in order to take a journey on how we could monitor and govern the whole process of deploying models and also monitoring how the models are performing across clients so there is where we went into partnership or we went to this like also co-creating different components because as we know there are so many things that you cannot have out of a box but then teaming up together, then something amazing comes out of it.”