From Enterprise Engineer to AI Researcher
Four years ago, I joined Bank of America as a fresher. Earlier this month, I concluded that chapter. This fall, I'll be starting an M.S. in Artificial Intelligence at Columbia University.
From the outside, that might look like a clean career pivot. From the inside, it was anything but.
During my Computer Science degree at SRM IST Chennai, I was first introduced to machine learning through coursework, projects, peer discussions, and exploring research literature. That curiosity stayed with me long after graduation.
But when I joined Bank of America in 2022, machine learning wasn't part of my day-to-day work. I was building Java microservices for corporate banking workflows. Debugging SSL expirations at 3 AM. Learning what it actually means to own a service in production — not just write it, but deploy it, monitor it, and be the person who picks up the phone when it breaks.
Over four years, I grew from an apprentice into the Subject Matter Expert for CashPro Search. I delivered 17+ production releases across 13 services. I resolved 50+ production incidents. I earned Gold, Silver, and Bronze Global Recognition Awards. I was named as an inventor on a patent for a governed multi-modal AI architecture.
That work taught me something about engineering that coursework can't: the difference between software that works and software that keeps working. Correctness under uncertainty. Systems that degrade gracefully instead of failing silently. The discipline of testing not just whether something is right, but whether it stays right.
Outside of work, the curiosity about AI kept pulling me back.
I started with Kaggle — publishing notebooks, climbing to Expert rank, eventually reaching a personal best of #707. But over time, my interest shifted from applying ML techniques to understanding how these systems behave.
The more I explored, the more I wanted to move beyond simply using modern AI systems. I wanted to understand how they learn, why they behave the way they do, and how to build them more reliably.
That led to a year of evenings spent building: LLM evaluation infrastructure, model governance pipelines, adversarial testing harnesses. The AI Reliability Platform. The drift-aware fraud detection system. Each project started as curiosity and evolved into something that felt like the beginning of a research question.
When I look back at Bank of America, what I'll remember most isn't a particular release, project, or technology. It's the people. The managers and teammates who trusted me with responsibility early, challenged my thinking, helped me grow through mistakes, and supported me throughout the journey. Every interaction, every challenge, and every lesson helped shape the engineer I am today.
At Columbia, I'm looking forward to combining the engineering discipline I learned in industry with a deeper understanding of modern AI. Not just how to build AI systems, but how to understand them — how they learn, how they reason, and how they can be made more reliable.
The curiosity that started during undergrad, survived four years of production engineering, and grew into hands-on AI projects is now becoming the foundation for graduate research.
Here's to the next chapter.