Intellectual Property and Confidentiality
As a consultancy business, it is critical for customers to trust us with their data and ideas, and we shall strive to not do anything which might undermine this trust.
All IP created during a consultation project will belong to the client. This means that all patents originating from our work belong to the client. Researchers from Rashmirathi Photonics may be listed as co-inventors on the patent application if they have made intellectual contributions to the development of the idea.
All the data provided by the client, and ideas developed together over the course of the collaboration shall be treated as confidential and will not be shared with any third parties.
Projects Related to Artificial Intelligence (AI)
My view is that the pinnacle of engineering is to transcend nature’s constraints in an elegant and resource-efficient way which is based on a deep understanding of physical first principles.
The philosophy of AI is in polar opposition to such values. Rather than transcending limits in an elegant way, it consumes huge amounts of resources to produce results whose accuracy is questionable. “Solutions” provided by black boxes do not provide genuine insight.
This makes any projects related to AI incompatible with my values as an engineer and as a human being.Therefore, Rashmirathi Photonics shall not take on any projects explicitly relating to AI, such as photonics accelerators for AI, designing components for AI computations, or consult for any company whose primary focus is on AI hardware and applications.
Furthermore, we shall not use AI in any of our R&D workflows. Every idea we produce and every action we take shall be directed by human intention and intuition. The aim is to make Rashmirathi Photonics into a “cleanroom” for intuition, and AI is one type of spiritual contaminant.
Music is more than mere sounds, art is more than just pictures, and literature is more than words strung together. All true art is based on individual human experiences and personal reflection on those experiences, which any AI is fundamentally incapable of doing. All “art” and “knowledge” created by AI is nothing but noise—a cheap and empty imitation of the real thing.
Gurukul System of Mentorship
For advanced training in high-skill fields, no system works better than the master-apprentice model of mentorship. I have personally benefitted from such a system for the past 20 years in diverse areas such as sports, engineering, and research, and now it is time for me to pass it on to the next generation. A little bit of individualized attention at the right time can go a long way in unlocking latent potential.
At Rashmirathi Photonics, I shall have only one or two early-career research assistants at a time for a period of 3-4 years each. Each assistant shall be individually trained by me in a manner tailored to their temperament, skills, and aptitude.
This method of training also benefits the client by having motivated and self-driven researchers working on their problems. The solutions created by people in a gurukul environment have a qualitative difference to them compared to solutions created by people who feel like replaceable cogs in an impersonal machine.
Academic Publishing
As a researcher to my core, publishing papers is how I built my career and motivated myself. Much of my own R&D accomplishments, through which I can now contribute to industry, would not have been possible without the published ideas and insights of others. As users of research published by others, we feel an obligation to give back to the semiconductor photonics community by sharing our own ideas and discoveries through the established medium of peer-reviewed journal publications.
I have extensive firsthand experience in navigating the delicate balance which must be struck between protecting a corporation’s trade secrets and sharing ideas to help the field progress.
When it comes to publishing, the key to a win-win cooperation between companies and researchers is to distill certain results to fundamental physics or to showcase record-performance engineering, while scrubbing sensitive design details not required to illustrate the key ideas.
Publishing papers can benefit all parties:
- The careers and motivation of individual researchers on both the client and consultancy sides
- The client organization, by marketing their technical competence, R&D culture, and contribution to the broader scientific community
In the end, the client has the final say on whether the work may be submitted to a peer-reviewed journal.
