Interview: Sanjeet Dutta, Data Scientist, BCG X, New York, United States

An interview with Sanjeet Dutta, Data Scientist, BCG X, New York, United States

Welcome to an insightful interview with Sanjeet Dutta, a distinguished Data Scientist at BCG X, located in the vibrant city of New York, United States.

With his extensive expertise in data analytics and machine learning, Sanjeet has been instrumental in driving innovation and strategic decision-making in various projects. Let’s delve into his journey, experiences, and valuable insights into the dynamic field of data science.

Can you walk us through your journey to becoming a data scientist, particularly highlighting your educational background and relevant work experiences?

Sanjeet Dutta: I was first introduced to Machine Learning during a program at Stanford in 2017 where I learnt about foundational Data Science concepts, statistics and technology entrepreneurship. This program piqued my interest enough for me to dedicate time for this domain beyond my undergraduate program’s commitments.

I started off by taking online courses ranging from foundation machine learning algorithms to advanced neural network architectures. During this time, I founded a club in my University for AI called Neuron where I built a community of driven individuals interested in the field.

To accelerate my growth, I also started working as a research intern for a prominent professor and researcher in IIM Bangalore on applications of Machine learning in Portfolio selection. This was a major boost for me since it cultivated the mindset of tackling hard research problems that require critical analysis.

While I continued working on this project throughout my undergraduate program, I also opted for industry experience through internships at the end of each academic year, often continuing this work into the start of the subsequent semester.

In the span of those 3 years, I had worked with computer vision, natural language processing, recommender systems and reinforcement learning models which helped me develop a wide range of knowledge across the AI domain.

After I joined Cornell for my graduate program, I continued to upskill myself in this field, picking this up full time in my current role. I owe my skills and expertise to the work I put in during my academic programs to develop these foundational skills.

What inspired you to pursue a career in data science, and how do you stay updated with the latest trends and developments in the field?

Sanjeet Dutta: As a Computer Science undergraduate student, I was curious to learn about emerging technology trends such as AI. I was amazed by the power of the algorithms that were coming up during the time such as facial recognition, sentiment analysis and predictive analysis.

The idea that real world phenomena could be represented in 1s and 0s was fascinating.  I was keen on learning how this “magic” really worked and once I built my first “hello world” project in house price prediction, I knew I was interested in taking this further through internships and research.

The advancements in AI have come a long way since the first feedforward neural network I built in 2017. As a data scientist, I actively seek out updates in the field through research publications, model releases and discussion forums to stay up to the date with the progress in the field. This is a crucial part of my role given the pace at which the field is evolving.

Could you share a challenging project you’ve worked on recently as a data scientist at BCG X? What were the key challenges you faced, and how did you overcome them?

Sanjeet Dutta: While I cannot speak to specific projects within my firm, I can comment on a key technology challenge which are industry wide . Navigating the advancements in Generative AI models over the past year has underscored the importance of thoughtful design and strategic implementation in technology solutions.

Given the non-deterministic nature of Generative AI outputs, establishing robust guardrails has been top of mind to ensure their reliability and applicability. This involves a careful balance between leveraging the potential of these models for innovative solutions and mitigating risks associated with their unpredictability.

Additionally, adapting to these advancements requires rethinking architectural design patterns to meet scalability demands effectively. My focus has been on integrating the latest research insights, community knowledge, and best practices to design state-of-the-art Generative AI applications that are not only scalable but also secured with necessary risk mitigation measures.

This strategic approach ensures that the solutions I develop are both cutting-edge and reliably grounded in the latest advancements in AI technology.

What programming languages and tools are you proficient in, and how do you leverage them in your daily work as a data scientist?

Sanjeet Dutta: In my journey as a data scientist, I’ve had the opportunity to work with a variety of programming languages, but my daily work primarily revolves around Python, SQL, and Node.js. Python, with its extensive libraries and community support, serves as my go-to language for data analysis and machine learning tasks.

It’s where I leverage powerful frameworks like PyTorch, TensorFlow, and Jax to build and iterate on machine learning models effectively. SQL is indispensable for data manipulation and extraction.

Node.js comes into play when I’m working on building or interacting with web applications, especially when setting up servers or APIs for data ingestion or model deployment. For cloud-based applications and data orchestration, AWS and Azure are my platforms of choice.

I also use Airflow frequently for orchestrating data workflows, ensuring that data processing and machine learning tasks are streamlined and automated.

BCG X likely deals with a variety of industries and sectors. How do you adapt your data science skills and methodologies to address the unique challenges and requirements of different projects and clients?

Sanjeet Dutta: In my view, data science skills are incredibly versatile; they’re tools that you can tailor to fit the task at hand. Successfully applying data science across different domains depends on learning fundamentals and really understanding the context.

In general, it is recommended to quickly get to grips with the domain—seeing beyond the data to what it actually means in that particular setting. The ability to adapt and learn fast matters more than sticking to a one-size-fits-all approach. It’s about shaping your data science toolkit to meet the specific challenges and needs of each project.

Data privacy and ethical considerations are increasingly important in data science. How do you ensure that your data analysis and models adhere to ethical guidelines and respect user privacy?

Sanjeet Dutta: Prioritizing data privacy and ethical considerations is non-negotiable. It starts with a solid understanding of both the legal requirements, like GDPR or CCPA, and the ethical standards that govern data use in specific contexts.

I ensure that data handling practices are transparent and consent-based, keeping stakeholders informed about what data is used and how it is collected, used, and shared. This involves conducting thorough impact assessments to identify and mitigate any potential risks to privacy or fairness.

Regular reviews and updates are part of the process, ensuring compliance with evolving standards and regulations.

Where do you see the future of data science heading, particularly in the context of consulting firms like BCG X, and how do you envision yourself contributing to that future?

Sanjeet Dutta: Happy to speak on my personal view on how Data Science will involve generally. I believe organizations are moving towards a more integrated approach where data science not only informs decisions but also drives innovation and operational efficiency.

This evolution will see a deeper reliance on predictive analytics, AI automation, and the development of personalized experiences, transforming the way businesses interact with their data and, by extension, their customers.

One of the key trends will be the democratization of data science, making powerful analytical tools and insights accessible to a wider range of users and decision-makers. This shift will require data scientists to focus not only on technical expertise but also on enhancing their ability to communicate complex data concepts in more intuitive ways.

As data science becomes increasingly fundamental to organizational strategy, the emphasis will also grow on ethical AI, privacy preservation, and the responsible use of data. Looking ahead, I see myself contributing to this dynamic field by staying at the cutting edge of technology, exploring innovative applications.

Our conversation with Sanjeet Dutta has shed light on the multifaceted role of data scientists in today’s rapidly evolving business landscape.

From leveraging advanced analytics to extract actionable insights to fostering collaboration across diverse teams, Sanjeet’s experiences exemplify the pivotal role of data science in driving organizational success.

We extend our gratitude to Sanjeet for sharing his expertise and wish him continued success in his endeavors at BCG X.

Read more Success stories of Indian entrepreneurs, Women Entrepreneurs & startups stories at SugerMint. Follow us on Twitter, Instagram, Facebook, LinkedIn