Vishwas Sharma · Specialist Data & AI
I didn't study industrial AI. I shipped it.
- M.Sc. Autonomous Systems, Bonn-Aachen International Center for Information Technology (b-it) — Robotics & AI (2022)
- Anomaly detection & Explainable AI for satellite operations — German Aerospace Center (DLR)
- Weightless neural networks on GPU for video background subtraction — Fraunhofer FKIE
- Established and led the AI team for global OEM projects at a Tier-1 automotive manufacturer
The journey
- 2013
Assistant Manager — Vedanta Ltd., Odisha
Started where most readers of this page are standing right now: an Indian engineering degree and no idea what the industry actually wanted. Commissioning and startup of a green anode plant, electrical and instrumentation. Heavy industry first, software later — which is why the projects here start from a physical problem.
- 2015
Electronics instructor — technical training institute, Bikaner
Led curriculum design and instructor training. First proof that the bottleneck in Indian technical education is not talent; it is what students are asked to build.
- 2019
Research Assistant — Fraunhofer FKIE, Wachtberg
Developed a CUDA-based unsupervised weightless neural network for video background adaptation — the work that became the BEWiS case in the showcase.
- 2021
Data Science Student — German Aerospace Center (DLR), Köln
Built an anomaly detection and Explainable AI system for satellite telemetry, aimed at catching the failure precursors that lead to loss of orbit.
- 2022
M.Sc. Autonomous Systems, Bonn-Aachen International Center for Information Technology (b-it)
Robotics & AI. Trained in the German research-institute way of working: real problems, rigorous validation, systems that ship.
- 2022
Specialist Data & AI — global Tier-1 automotive manufacturer, Bonn
Established and led an AI team for global OEM projects: 3D generative AI for shape optimisation, a vision system for quality control, active learning to cut the number of physical experiments, and the design–material–process work that became the Physical AI Platform.
- 2024
Kautex Innovation Award · Textron Chairman’s Award finalist
Personal awards for the innovation work, alongside leading the digitalisation initiative for strategic business reviews and presenting to C-suite executives.
- 2026
The Industry-Incubator, Jaipur
Home, with one conviction: the way German institutes build engineers can be built here — real problems, small cohorts, public proof.
Why I refuse to teach AI the way EdTech sells it.
“AI's real value isn't in its ability to predict — it's in how it drives faster, smarter, and more consistent decisions.”
So apprentices are never graded on model accuracy alone. Every project must end in a decision a real engineer or operator can act on — that is the bar the lab holds work to.
“A well-generalized model with 70% accuracy that works in the real world is infinitely more valuable than a 98% accurate model that can’t be used.”
That is why the lab ships deployed systems, not notebooks. Apprentices learn the unglamorous parts — data quality, validation, serving, monitoring — because that is where industrial AI lives or dies.
“The question isn't what AI can do — it's what critical decisions we need it to support.”
Every cohort project starts from an industrial decision that matters, not from a technique looking for a use case. The problem statement comes first; the syllabus is whatever it takes to solve it.
What I will not promise you.
I won't promise a guaranteed job. I won't show you fake placement percentages. I promise real problems, German-research-grade mentorship, and a portfolio that speaks in interviews. What you do with it is yours.