Best Place To BuildProf. Prabhu Rajagopal l"Brain drain isn't about salary. We want to be challenged"| Ep. 3
CHAPTERS
IIT Madras as a “best place to build”: meeting at the Sudha & Shankar Innovation Hub
The host sets the premise of the series—understanding why IIT Madras is considered a great place to build—and introduces Prof. Prabhu Rajagopal at the Innovation Hub. They establish their shared background as Mechanical Engineering / Intelligent Manufacturing alumni and tee up themes of innovation culture and startups.
- •On-location setup at Sudha & Shankar Innovation Hub
- •Goal of the podcast: learn from builders and identify what enables building at IITM
- •Introduction of Prof. Prabhu Rajagopal and his connection to the host (same program)
- •Early hint: IITM’s ecosystem is increasingly startup-oriented
From “Intelligent Manufacturing” to Industry 4.0: how the curriculum anticipated the future
Prabhu explains what “intelligent manufacturing” meant in the late 90s/early 2000s and how it maps to today’s Industry 4.0 reality. The conversation frames intelligent manufacturing as sensing, feedback, connectivity, and machine-to-machine intelligence—now commonplace on modern shop floors.
- •Early vision: sensors embedded into manufacturing processes
- •Live feedback loops to optimize cutting/feed parameters and operations
- •Connected, networked machines communicating with each other
- •Industry 4.0 as the mainstream evolution of those ideas
Bhatnagar Prize recognition: personal roots and “translational” impact
He recounts winning the Shanti Swarup Bhatnagar Prize and why it’s personally meaningful due to his father’s CSIR background. He emphasizes that the award (technology & innovation category) validates translational work—moving beyond papers into field deployments through industry and startups.
- •Personal significance: CSIR roots and family pride
- •Historical context: luminaries among past awardees
- •Award category: technology and innovation (not just academic publication)
- •Focus on translating research into real-world use via industry/startups
What “science administration” means: building mission-oriented innovation capacity
Prabhu defines science administration as creating structures that enable mission-critical science and technology to scale—analogous (in spirit) to institution-building by Bhabha or Sarabhai. At IITM, this shows up as creating programs and platforms (CFI/NIRMAN) that nurture student innovation into products and ventures.
- •Science administration as “institution-building” for impact
- •Administrative push to enable mission-oriented R&D
- •CFI/NIRMAN as mechanisms that nurture innovation and product creation
- •Supporting students from ideas to solutions, startups, and deployments
The IIT Madras innovation stack: CFI → NIRMAN → Incubation (and research commercialization)
Using a whiteboard, Prabhu lays out IITM’s innovation pipeline and how students and researchers flow through it. He describes CFI as the practical making core, NIRMAN as a rare pre-incubator, and other feeders (E-Cell/TechSoc/GDC) that connect student projects and faculty research to commercialization and incubation.
- •CFI as the heart of hands-on, interdisciplinary building (cars, rockets, Hyperloop)
- •IC&SR as a long-standing backbone enabling industry interaction/support
- •Feeder orgs: TechSoc, E-Cell; commercialization path via GDC
- •NIRMAN as pre-incubator; incubation cell and sector incubators downstream
- •Shift: MS/PhD founders increasingly use NIRMAN (100+ pre-incubated startups)
Changing academic culture and “brain drain”: it’s not salary, it’s challenge
They contrast the earlier default paths (government roles, IT services, or going abroad) with today’s broader opportunity set in India. Prabhu argues brain drain discussions missed a key point: many people leave (or stay) based on access to challenging, high-impact problems—not only compensation.
- •Earlier era: many graduates went abroad; fewer local “core” options
- •Today: more people stay as opportunities have expanded
- •Brain drain reframed: desire for challenge and meaningful problems
- •PhD motivations changing: startup/CTO aspirations now common in interviews
Startup origins #1: Planys—underwater robotics for inspection and infrastructure health
Prabhu narrates how Planys emerged from CFI club work (AUV/ROV teams) and a student’s master’s project, with industry pull from Reliance helping catalyze the team. The company’s applications span oil & gas tanks, bridges, and dam inspections, evolving into a globally operating pioneer in Indian underwater robotics.
- •CFI roots: robo-sub team, Robocon club, master’s project in pipeline inspection
- •Industry validation and early support (e.g., Reliance interest)
- •Applications: tank inspection, bridge river crossings, dams/hydel infrastructure
- •Growth: first Indian underwater robotic startup; international operations/subsidiary
Structural Health Monitoring (SHM) and the “Internet of Underwater Things”
The discussion shifts from the startup story to the underlying engineering philosophy: monitoring infrastructure while it operates. Prabhu explains underwater constraints (no normal internet) and how Planys uses acoustic communication plus surface/satellite relays to enable persistent sensing and data collection.
- •SHM as continuous monitoring of aging infrastructure under operation
- •Underwater comms limitations addressed via acoustics-based systems
- •Acoustic beacons + AUV/ROV flybys + surface/satellite data transmission
- •Underwater SHM as a differentiator for inspection at depth
Startup origins #2 and NDE foundations: Solinas and the role of non-destructive evaluation
Prabhu links Solinas to a dual-degree project and highlights its impact on water/sewer networks and eliminating manual scavenging. He then explains NDE as “seeing inside” structures without damaging them—drawing parallels with medical X-rays/ultrasound—and how robotics becomes a carrier for NDE sensing in hard environments.
- •Solinas roots: student project (Homo Sap robot) → city-scale deployments
- •Focus areas: water/sewer maintenance; reducing manual scavenging
- •NDE explained: inspect without damage; prevent catastrophic failures
- •Tech examples: underwater ultrasonics, laser-based internal pipe inspection
Startup #3: Xyma and waveguide/fiber acoustics for harsh industrial sensing
Prabhu describes Xyma as deep-tech commercialization of waveguide acoustics—sending sound through a “wire” (waveguide) to measure properties in extreme temperatures. The approach enables remote inference of temperature/viscosity/rheology in processes like smelting and refining, where conventional sensors fail.
- •Waveguide concept: transmit acoustic waves through inexpensive/robust materials
- •High-temperature sensing in smelting, refining, nuclear contexts
- •What’s inferred: temperature, viscosity, rheology, process state
- •Industry adoption across oil & gas, mining, petrochemicals, paints/lubricants
- •Ongoing frontier: material limits for very high temperatures (e.g., steel)
TRLs and the lab-to-field “valley of death”: why startups become the missing bridge
Using Technology Readiness Levels, Prabhu explains how academia traditionally stops at TRL 1–3 (proof of concept), while field deployment needs TRL 7–9. He positions CNDE as the R&D/IP engine and startups as the commercialization arm that carries solutions through the valley of death, leveraging industry problem pipelines.
- •TRL breakdown: 1–3 proof of concept; 4–6 mockups; 7–9 field studies
- •Traditional academic stopping point creates a “valley of death”
- •CNDE scale: large team across PhDs, masters, and project staff
- •Industry provides real problem statements; lab develops deep tech; startups deploy
CNDE startup constellation: multiple ventures, AI-driven inspection, and data scale-up
Prabhu lists the broader CNDE startup ecosystem (beyond Planys/Solinas/Xyma), showing a pipeline of adjacent companies across inspection, drones, and operations. He then explains the data explosion from robots/sensors and the shift from manual analysis to AI summarization and feature extraction.
- •Additional ventures: Dhvani (split into multiple companies), Detec, Maximal Labs, more
- •Trajectory: pipeline inspection → drones → occupational health and analytics
- •Robotics produces GB/TB-scale inspection data; manual review doesn’t scale
- •AI now summarizes/analyzes inspection outputs; custom algorithms built in-house
Why blockchain matters for AI: data fidelity, privacy, and healthcare interoperability (Plenome)
Prabhu argues AI’s promise depends on trusted, untampered data and privacy-preserving access—areas where blockchain can help through provenance and anonymization. He connects this to COVID-era pain points in medical data portability and introduces Plenome, focused on cybersecurity-backed interoperability in healthcare systems.
- •Blockchain as a trust layer for data authenticity/provenance
- •Privacy need: anonymization and controlled access to sensitive data
- •COVID highlighted interoperability gaps across labs/states/geographies
- •Plenome’s focus: healthcare data security + interoperability as an outcome
Cross-disciplinary engineering mindset: beyond departmental silos, toward “general engineering”
Responding to concerns about straying beyond mechanical engineering, Prabhu frames modern engineering as inherently cross-disciplinary. He positions deep domain expertise as valuable, but notes that at sufficient depth, fields converge—especially through data, computation, and pattern recognition.
- •NDE spans mechanics, electronics, instrumentation, signal processing, AI
- •Cross-disciplinarity intensified by democratized computing/tools
- •Discipline matters for deep domain grounding, but convergence happens at depth
- •AI/data science as unifying layers across domains
Student experimentation to real deployments: blockchain elections and campus-to-startup pathways
He describes IITM’s WebOps/Blockchain club projects, especially conducting distributed-node blockchain elections using student ID authentication. The elections moved from a club project to repeated execution and later became part of the startup’s applied work—illustrating IITM’s “play → build → productize” loop.
- •Blockchain club as a maker community focused on trust systems
- •Elections as a “hard” trust problem; blockchain as distributed verification
- •Campus implementation: authentication + distributed nodes + counting/results
- •Pattern repeated: student project → institutional use → startup execution
Life at IIT Madras: nicknames, alumni bonds, Vivekananda Study Circle, and intellectual roots
The conversation becomes personal: his IIT nickname (“Gandhi”), the campus nickname culture, and the enduring alumni network. He also discusses the Vivekananda Study Circle as a venue to explore Indian philosophy and identity, alongside formative coursework that encouraged multiple historical narratives.
- •Nickname culture as a distinctive IIT social tradition
- •Alumni support network across geographies (hosting/meeting while traveling)
- •Vivekananda Study Circle: engagement with Indian/Hindu philosophy and texts
- •Value of multiple narratives and primary sources in understanding history
Poetry, philosophy, and motivation—plus new ventures and the IITM “best place to build” formula
Prabhu connects poetry to a broader search for beauty and truth in work, linking it to philosophical ideas like Satyam-Shivam-Sundaram. He then outlines newer startup directions (AI for 3D printing, AR/VR representations, foundational robotics, ML water metering) and closes with why IITM excels: industry access, making culture, and a startup culture supported by a mature stack.
- •Poetry as a philosophy of finding beauty/meaning across work
- •New startups: Matterize (AI for 3D printing), Zeroscape (AR/VR), Botforge (robot models), TerraClime (ML water metering)
- •Startups still “fringe” for most students; goal is to mainstream entrepreneurship
- •IITM’s differentiators: strong industry interactions + student making + startup culture
- •Examples of ecosystem successes: Agnikul, Galaxeye, Ather, Modular housing; long-run institutional maturity (IC&SR/CFI/NIRMAN/Incubation)
