
Prof. Krishnan B.| He X-Rays Bridges & Planes | He Left USA to Build Non-Destructive Testing | Ep. 8
Krishnan Balasubramanian (guest)
In this episode of Best Place To Build, featuring Krishnan Balasubramanian, Prof. Krishnan B.| He X-Rays Bridges & Planes | He Left USA to Build Non-Destructive Testing | Ep. 8 explores how IIT Madras scaled deep-tech NDT into startups and impact Non-destructive testing (NDT) is positioned as “engineering diagnostics,” using tools like X-ray and ultrasound to detect internal defects and prevent catastrophic failures in assets from trains to pipelines.
How IIT Madras scaled deep-tech NDT into startups and impact
Non-destructive testing (NDT) is positioned as “engineering diagnostics,” using tools like X-ray and ultrasound to detect internal defects and prevent catastrophic failures in assets from trains to pipelines.
Balasubramanian recounts returning from the US in 2000 to build IIT Madras’ Center for NDE from scratch, quickly attracting strategic-sector and industrial support and scaling it into a high-impact capability.
As Dean (ICSR), he describes ICSR as IIT Madras’ industry-facing window and highlights process reforms that accelerated IP protection and increased patent filings dramatically while preserving academic publication velocity.
He details how IIT Madras’ incubation policy and institutional structures (Incubation Cell under Research Park, CFI, Nirmaan, Research Park) made commercialization repeatable rather than ad hoc.
Multiple startups (Dhvani, Planys, Detect, Xyma) are presented as case studies in “hard problem solutions,” combining hardware, sensing, robotics, and AI to create durable moats and global customers.
Key Takeaways
NDT is the maintenance-and-safety backbone of modern infrastructure.
Balasubramanian frames NDT as the equivalent of medical imaging for engineered assets, enabling early detection so repairs happen during downtime, improving both safety and availability.
Get the full analysis with uListen AI
Materials and manufacturing evolve faster than biology—so inspection must keep evolving.
Unlike human anatomy, engineered materials, weld processes, and designs change frequently; this makes NDT a continuously moving technical target rather than a solved discipline.
Get the full analysis with uListen AI
Quality assurance is needed both at manufacturing and throughout service life.
He distinguishes “as-built” defect detection (e. ...
Get the full analysis with uListen AI
Fast, faculty-friendly IP processes can shift a campus from publishing-only to impact-driven.
ICSR reforms aimed to protect inventions without slowing papers; he cites reducing patent-processing timelines (down to ~24 hours in exceptional cases) and scaling filings to hundreds per year.
Get the full analysis with uListen AI
Incubation succeeds when governance, legal structure, and handoffs are engineered—not improvised.
IITM’s approach navigated Board approvals and state legal constraints (Tamil Nadu Society Act), ultimately using a Section 8 structure under Research Park to enable equity and proper incubation.
Get the full analysis with uListen AI
Deep-tech startups often start as industry pain points plus a lab advantage, then pivot to platforms.
Detect began with a high-temperature corrosion sensor, expanded into drone-based inspection, and then evolved into a broader AI platform for asset integrity, compliance, and human safety management.
Get the full analysis with uListen AI
Entrepreneurship can be taught as a mindset shift from “my invention” to “customer discovery.”
The I-Corps program is described as a structured process that forces teams to validate who will buy, why they won’t, and what must change—effectively “curating entrepreneurs,” including faculty founders.
Get the full analysis with uListen AI
Notable Quotes
“We look inside materials… just the way a diagnostic center… looks inside your anatomy.”
— Krishnan Balasubramanian
“[On welding assurance] …his answer he gave me was one word. He said, “Entropy”… And his answer is, “Probably never.””
— Krishnan Balasubramanian
“I landed here with nothing… except my suitcase… And then they gave me a small office and said, “Do something.””
— Krishnan Balasubramanian
“We are just not curating ideas, we are curating entrepreneurs.”
— Krishnan Balasubramanian
“Unlike X, SpaceX is not going to go away that easily… when you have deep tech… [it] will keep growing.”
— Krishnan Balasubramanian
Questions Answered in This Episode
On NDT scope: Where is the practical boundary between ‘detectable defect’ and ‘acceptable defect’ across materials like ceramics, metals, and concrete?
Non-destructive testing (NDT) is positioned as “engineering diagnostics,” using tools like X-ray and ultrasound to detect internal defects and prevent catastrophic failures in assets from trains to pipelines.
Get the full analysis with uListen AI
On welding: If welding quality can ‘probably never’ be guaranteed due to entropy, what inspection-plus-process controls best reduce real-world failure rates?
Balasubramanian recounts returning from the US in 2000 to build IIT Madras’ Center for NDE from scratch, quickly attracting strategic-sector and industrial support and scaling it into a high-impact capability.
Get the full analysis with uListen AI
On corrosion economics: You cite corrosion as a ~$5 trillion global problem—what interventions (inspection frequency, coatings, materials choices) deliver the biggest ROI in India’s infrastructure?
As Dean (ICSR), he describes ICSR as IIT Madras’ industry-facing window and highlights process reforms that accelerated IP protection and increased patent filings dramatically while preserving academic publication velocity.
Get the full analysis with uListen AI
On IP execution: What specific workflow changes enabled IITM to move from ~20 patents/year to ~400/year, and what trade-offs did you have to manage?
He details how IIT Madras’ incubation policy and institutional structures (Incubation Cell under Research Park, CFI, Nirmaan, Research Park) made commercialization repeatable rather than ad hoc.
Get the full analysis with uListen AI
On incubation governance: How did IITM convince a diverse Board of Governors to approve incubation, and what were the key guardrails to address conflict-of-interest concerns for faculty founders?
Multiple startups (Dhvani, Planys, Detect, Xyma) are presented as case studies in “hard problem solutions,” combining hardware, sensing, robotics, and AI to create durable moats and global customers.
Get the full analysis with uListen AI
Transcript Preview
Unlike X, SpaceX is not going to go away that easily, okay? If anything, Space- SpaceX will keep growing. That's my belief, that when you have deep tech... So we are just not curating ideas, we are curating entrepreneurs.
Right.
We are curating faculty who become entrepreneurs. Uh, Reliance actually say, "I'm, uh, going to actually support this work, and what- whatever we help you productize it, and we want you to try it out on our... And we will also buy 200 of these sensors if it works." [upbeat music]
Hi, my name is Amrit. We've heard that IIT Madras is the best place to build. [upbeat music] So we've come down to the Sudha and Shankar Innovation Hub. We want to meet some people. These are builders. We want to talk to them about their work, and also ask them, "What makes IIT Madras the best place to build?" [upbeat music] Hi, welcome to the Best Place to Build podcast. We are sitting today with Professor Krishnan Balasubramanian, who is a professor at, in mechanical engineering at IIT Madras. He's a very well-accomplished and very well-loved professor, award-winning professor. Recently, uh, in 2017, won the Abdul Kalam National Innovation Fellowship. Uh, welcome to the podcast, Professor.
Thank you, Amritash.
Professor-
Very happy to be here.
Uh, and, uh, also, I might just start off by saying that I was a student of yours in, in your class in 2003, '4, uh, first year, second year. So I have memories from the class also. So really happy that I'm meeting you again after so many years. Uh, Professor, you've had a long, um, career in IIT Madras, and you're obviously associated with a lot of things. But let's start from the beginning. Uh, your- can you talk to us about your research area and, uh, the work that you do as a pr- as a faculty member?
Thank you, Amritash. Um, it's, it, it looks, feels lo- uh, very short, but it's probably 24, 5... It's getting to be close to 25 years since I returned back from US and joined IIT Madras. Um, my area of expertise is in an area called as non-destructive testing, which, to many of you, may be foreign language. But essentially, you all remember that whenever you have anything to do with your health, you go to a doctor, and then the doctor says, "Go, go get an ultrasound done. Go get an X-ray done." What you do is you look inside your anatomy, and you're able to find out if there everything is okay or if something is not okay. Now, imagine the number of the, uh, you know, engineered, uh, structures and components that you use every day, whether it's a car, so you're taking a flight, you are in a train, you're using the rail, you're using the bridge, crossing a bridge. You have pipelines, you know, transporting gas to your h- home. You have, uh, sewage lines. Each of them are all good when you're building it, and then over a period of times, things start to go wrong. And when things go wrong, either you- things become unavailable to you. Imagine suddenly that you have a flight that's canceled because of some maintenance issue. Now, if I can prevent that by letting you know that this c- plane has a problem, and then when it's not being used, you fix it, then it becomes available for you every time. Of course, you can also look at it from a different perspective, that we are looking at safety. So if you look at the number of catastrophic incidences that happen, uh, if I can prevent many of them, then that is a mission that we are embarked on. So we look inside materials, we look inside components, engineered component, we look inside structures, just the way a diagnostic center just down the road from you looks inside your anatomy and says whether you have a valve that is not working or whether it's a heart that is, you know, blocked. Exactly the same thing we do, but big difference is that your anatomy has not changed in zillion years. On the contrary, engineered materials change almost every year. Components are, I guess, redesigned every couple of years. So our problem and challenges are evolving on a daily basis, and we have to keep solving newer and newer problem because we have new materials, new problems associated with new materials. I can give you an very- one good example. Um, I remember many years ago, the railways came to me and says, told me that the wheels are cracking up, and there have been accidents because of that. And the reason why that happened was because they changed the brake pad from a metal brake pad to a composite brake pad. Why did they do that? They did that because composite brake pads actually stop your train much closer. So within about, uh, 500 meters, you could stop the train, before the train was, would take two me- kilometers to stop. So that looked like a brilliant advantage. But what they didn't realize or, uh, or probably what the challenge here was, that the kinetic energy that you have has to be dissipated in 500 meters what was dissipated in two kilometers. So now you have heat that's heating up your wheels a lot faster, and then your metallurgy is not able to take it. So now they wanted to know whether I can predict which wheel is going to fail so that they can remove it out before anything bad happens. So-... we can do this by many of the NDT methods that are available today. So thank you.
Install uListen to search the full transcript and get AI-powered insights
Get Full TranscriptGet more from every podcast
AI summaries, searchable transcripts, and fact-checking. Free forever.
Add to Chrome