Deepak N. Subramani

About Me

Mon, 01 Jan 0001 00:00:00 +0000

deepakns@iisc.ac.in | (O): +91-80-2293 2058 | cds.iisc.ac.in/faculty/deepakns Address: Dept. of Computational and Data Sciences, Room No. 321, Indian Institute of Science, Bengaluru 560012

Experience

Indian Institute of Science, Bangalore, India

Associate Professor, Department of Computational and Data Sciences — QUEST Lab, Division of Interdisciplinary Sciences, IISc 31 Dec 2024 – Present

Agentic AI Coding: From Specification to Deployment

Mon, 01 Jan 0001 00:00:00 +0000

A hands-on 6-weekend program for professionals and students who want to build, ship, and operate real AI agents — from writing a spec to deploying a full-stack application. Launches September 5, 2026, via IISc’s Centre for Continuing Education.

Contact

Mon, 01 Jan 0001 00:00:00 +0000

Reach out for collaborations, media inquiries, or anything else. Prospective students asking about internships or positions should check Join the Lab first — it covers exactly what makes an inquiry email work.

Industry Relations

Mon, 01 Jan 0001 00:00:00 +0000

Some of this research is funded to solve a specific industry problem, not just to publish. Below are the active and completed industry-sponsored projects where I’ve been Principal Investigator.

Shell India Technology Limited

WAKE-NET: A Fast and Scalable Physics-Informed Neural Network for Wind Farm Simulation — ₹1.92 crore, 2022–2025, PI. Built the first physics-informed neural network for simulating turbulent wake behind wind turbines, cutting simulation cost relative to traditional CFD. Published in Physics of Fluids.
AI for Heat Exchanger Application — ₹90 lakh, 2025–2026, PI. Ongoing.

ISRO NRSC

Development of a Machine Learning Tool for the Prediction of City-Scale Extreme Rainfall Events using a Hybrid Physics-AI Approach — ₹43.3 lakh, Mission Mode Project, 2025–2027, PI. Ongoing.

Altigreen Propulsion Private Limited

Deep Learning for EV Fault Detection — ₹11 lakh, 2022–2024, PI.

Looking to sponsor research or explore a collaboration? Get in touch.

Research

Mon, 01 Jan 0001 00:00:00 +0000

My research asks one question in seven different settings: how do you build machine learning that respects the physics, uncertainty, and constraints of the real world, instead of treating it as a black box? That question started during my PhD at MIT with Pierre Lermusiaux — stochastic optimal control and uncertainty quantification for ocean path planning, work that produced a US patent and, eventually, grew into the QUEST Lab’s current research program at IISc.