I’m Tanvi Tiwari, a final-year data science student at SP Jain Global, Sydney. When I joined this program, I thought data science was mostly about learning tools, building projects, and getting models to work. I didn’t realise that the real transformation would come when I started thinking like a researcher, not just a student.
In my second year, I studied Machine Learning with Dr Aditya Narvekar (Deputy Director of the Program). That class changed the way I understood the field. It wasn’t only about algorithms or improving accuracy. He taught us how to think clearly about the problem, question assumptions, and justify every step with evidence. For the first time, I saw data science as something deeper than “building models.” It was about building reasoning.
That shift stayed with me. Over time, that curiosity turned into consistency. I started taking research seriously because I realised coursework rewards correct answers, but research rewards better questions. In research, you don’t just apply a method, you defend it. You don’t just report results, you explain what they mean, what they depend on, and what they cannot claim. That process forced me to become more rigorous and more intentional with my work.
The most defining step in this journey was getting the opportunity to work under the mentorship of Dr Aditya Narvekar on his Springer Nature book chapter. The chapter is titled “Bankruptcy Prediction: Statistical Models to Deep Learning Models”, published in the book “Navigating AI in Business: Strategies and Insights Across Disciplines”. It explores a practical question: how different modelling families perform for bankruptcy prediction, and what we gain or lose as we move from classical statistical methods to modern deep learning. The trade-offs matter just as much as performance, including interpretability, stability, and whether the system can be trusted in real decision-making.
My contribution focused on supporting the research process. I worked on literature review and synthesis, helped organise findings into a coherent progression from traditional models to deep learning, and supported writing in a way that is clear and defensible. More importantly, I learned how to do research the right way. Prof Aditya emphasised structured reading, clean reasoning, and honest claims. Under him, I learned that research is not about sounding impressive, it is about being methodical and precise in your approach.
Publishing at this level didn’t make me feel “done.” It made me feel capable, because I learned a repeatable process that I can trust: read deeply, ask better questions, test honestly, write clearly, and iterate. It also clarified what kind of work I want to do next. I’m interested in building analytics and AI systems that people can trust, especially in environments where decisions have consequences. Accuracy matters, but reliability, transparency, and responsible data use matter just as much.
If you’re a student who wants to go from “learning data science” to actually contributing to it, my biggest advice is to start before you feel ready. Begin with one narrow question and go deep for two weeks. Write as you go because those notes become your research. Seek feedback early because mentorship accelerates growth. And don’t underestimate consistency, because small daily effort compounds into outcomes you can’t predict at the start.
Looking back, the biggest change wasn’t my technical skills. It was my mindset. I came in as a student trying to learn data science. Through mentorship, discipline, and research, I started learning how to create knowledge, not just consume it. And I’m grateful that the guidance I got at SP Jain Global played such a central role in that journey.
If you are interested in reading the Springer chapter, you can get the eBook at: https://link.springer.com/book/10.1007/978-3-032-02526-5.
Tanvi Tiwari is our final-year Bachelor of Data Science student, currently in Sydney.
Recommended Reads: