Hello NVIDIA Team,
I am a PhD Scholar at Amity University Patna and an AI/Generative AI educator, currently developing an academic semiconductor manufacturing digital twin using NVIDIA Omniverse, OpenUSD, MQTT, and GPU-accelerated simulation.
I am developing this project as part of my academic research and educational work. I have been building it incrementally and have now developed a substantial working OpenUSD and data-integration foundation.
The current prototype includes:
- USD-based semiconductor manufacturing cell
- Semiconductor equipment and material-handling models
- OpenUSD scene hierarchy and metadata
- MQTT-based live telemetry integration
- SQLite operational and production data
- OEE and KPI analytics
- What-if failure and recovery simulations
- Bottleneck identification
- Decision-support scenarios
- USD-based decision feedback
- End-to-end integration validation
- Omniverse deployment-readiness validation
The project has progressed through 17 development levels, with the current deployment-readiness validation passing for the OpenUSD/data side of the system.
I have now reached the stage where I need an NVIDIA RTX-capable GPU environment to validate the GPU-dependent portion of the project, including:
- NVIDIA Omniverse / Kit
- USD Composer
- RTX rendering
- PhysX simulation
- Sensor RTX
- Production CAD/SimReady asset workflows
- Eventually Nucleus-based collaboration
My personal laptop has an AMD Ryzen 3 3200U processor, 8 GB RAM, and integrated AMD Radeon Vega 3 graphics, so it cannot run the required RTX-based Omniverse workloads.
I am currently exploring NVIDIA Brev and found an L40S configuration that appears technically suitable for this research. However, the cloud GPU is billed hourly, which is difficult for me to afford as a PhD scholar.
I would therefore like to kindly ask whether NVIDIA currently offers any PhD/student GPU credits, educational Brev credits, free GPU credits, academic programs, or other resources that could support this research project.
Even a limited amount of GPU credit would be extremely valuable. I would use the resources strictly for academic research and educational purposes to complete and validate the Omniverse portion of this semiconductor digital-twin project.
As an Generative AI educator, I am also interested in using this work as a practical educational example of how NVIDIA technologies can be applied to semiconductor manufacturing and digital-twin research, connecting:
OpenUSD → Omniverse → RTX Visualization → Simulation → Live Telemetry → KPI/Decision Support
I have published GitHub repository containing the educational code, USD examples, architecture, and validation workflow. No proprietary, confidential, credential, or private company information will be included.
I would be very grateful for any information about available PhD/student GPU-credit programs, Brev educational credits, NVIDIA academic resources, or other opportunities that could help me continue this research.
I would also be happy to document the results, share the project outcomes, and provide feedback about using NVIDIA technologies for semiconductor digital-twin research and education.
Thank you very much for your time and for supporting students, researchers, educators, and developers.
Best regards,
Amit Kumar
PhD Scholar, Amity University Patna
Generative AI Educator
Semiconductor Manufacturing Digital Twin — Academic Research Project