His internationally published research, “Innovative Smart Water Management System Using Artificial Intelligence” (Turkish Journal of Computer and Mathematics Education, 2021), presents a low-cost, plug-and-play platform that uses IoT sensors, machine learning, and cloud analytics to monitor and predict household water consumption in real time. Early tests show that the system can achieve 95 percent leak-detection accuracy and reduce water wastage by up to 30 percent in controlled pilot environments.

“Drumil’s work exemplifies the kind of socially responsive innovation India needs,” says Dr. Sunil H. Karamchandani, Faculty, DJSCE Department of Electronics & Telecommunication. “He has combined advanced data science with practical engineering to solve a problem that affects millions. It is rare to see this level of technical maturity and civic awareness in an undergraduate.”

Q: Drumil, what inspired you to take on water management as your focus?

A: The idea came from noticing everyday waste, a dripping tap, an overflowing tank, and realizing how invisible it all was. During the lockdown, when I was spending more time at home, it struck me that we could use AI to make people aware of their consumption before it turned into waste. I wanted a system that any Indian household could afford and install without technical help. For me, engineering has always been about usefulness, not just theory.

Q: How does your system actually work?

A: The hardware revolves around an ESP32 microcontroller connected to ultrasonic water-level sensors. These sensors send data to Firebase Cloud, where our machine-learning pipeline processes it. We used multiple models: Linear Regression, Random Forest, and Gradient Boosting, but Gradient Boosting combined with SARIMAX time-series forecasting gave the most stable results, predicting usage up to five days ahead with a root-mean-square error below 0.1.

When the system detects an abnormal spike, say, from a leak, it pushes an alert to a mobile dashboard built on MIT App Inventor. Everything runs wirelessly and can be scaled for apartments, societies, or small towns. The aim was simplicity without sacrificing intelligence.

Q: Why is this particularly significant for India?

A: Because India’s water stress is more about management than availability. We lose billions of liters through leakage and unmonitored flow. Our system brings visibility. Once people see their consumption data, behavior changes immediately.

If deployed city-wide, the data can feed a smart-grid-style water network, letting utilities pinpoint losses or forecast shortages. It aligns perfectly with India’s Smart City Mission and Sustainable Development Goal 6 on clean water. We’re proving that affordable technology can make conservation measurable.

Q: What challenges did you face as a student researcher?

A: The biggest hurdle was access to real-world datasets and lab resources during the pandemic. We built our own miniature test-bed at home using recycled pipes and tanks. I also had to self-learn statistical modeling, going from YouTube tutorials to peer-reviewed papers overnight!

There were many failed prototypes. But with mentorship from Dr. Karamchandani and my co-authors, we refined the design, validated data integrity, and achieved consistent accuracy. That persistence taught me more than any textbook ever could.

Q: What’s the next step for your Smart Water project?

A: We’re working on municipal-scale pilots intending to connect several hundred sensors across residential blocks. We also plan to integrate AI-driven anomaly clustering, so the system can differentiate between leaks, seasonal variation, and legitimate consumption spikes.

Long-term, I envision an AI-enabled National Water Grid as an open-data platform where each home or plant becomes a node contributing to predictive analytics for the entire city or region. The same model could extend to agriculture, helping farmers optimize irrigation and save electricity used in pumping.

“The prototype has proven that scalable sustainability is achievable when young minds think systemically,” notes Prof. Ranjushree Pal, Faculty, DJSCE E&TC. “Drumil’s combination of technical depth, accuracy metrics, and community orientation positions him among the top one percent of innovators in his generation.”

Q: You’ve received international recognition for this research. What does that mean to you personally?

A: It’s deeply validating, but more than recognition, it’s a reminder that impact has no age bar. Knowing that our work was accepted in a global journal while I was still an undergraduate showed that ideas from India can compete internationally. My dream is to continue in AI for Sustainability, take this to industrial and governmental scales, and mentor other students who want to turn code into change.

Q: Finally, what message would you share with young engineers across India?

A: Start where you are. Every problem you notice traffic, power cuts, and waste, can become your innovation. Don’t chase glamour; chase relevance. Learn the math, understand the data, and build with empathy. Technology should not just predict the future; it should protect it.