Update README.md

README.md

@@ -146,3 +146,32 @@ Bergholm, V., et al. (2018). PennyLane: Automatic differentiation of hybrid quan
 Fortin, F. A., et al. (2012). DEAP: Evolutionary algorithms made easy. Journal of Machine Learning Research, 13(Jul), 2171-2175.
 
 Moritz, P., et al. (2018). Ray: A distributed framework for emerging AI applications. In 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18) (pp. 561-577).
+
+
+
+
+## Subject: Clarification and Reflection on the Contest
+
+Dear JoAnn,
+
+I hope this message finds you well. I want to start by thanking you for your time and attention in reviewing my case. My intention with this email is not to request any changes to the contest results, as winning was never my goal. My main interest was simply to showcase my research, papers, and patents to NVIDIA engineers, as I deeply admire your work.
+
+I have worked as a researcher for many years in the field of biotechnology, and you can find my patents on Google Patents: Francisco Angulo Lafuente - Patents. However, I see myself more as a science fiction writer of B-movie-style novels and as a programming and artificial intelligence enthusiast.
+
+I would also like to apologize because I understand that part of the misunderstanding may have arisen due to my limited command of English, as I am Spanish and speak only Castilian fluently. I am sure that, if I had been able to communicate better in English, this issue could have been resolved easily from the very beginning.
+
+After much reflection, I believe I now understand your perspective. The typical approach to integrating NVIDIA NeMo technologies would have been to download the GitHub repository and integrate it directly into my code. However, to do so, the code would need to be written in Python.
+
+My project, although it may not appear so at first glance, is more complex because it combines various NVIDIA technologies, such as CUDA, Raytracing, AI RAG, and NVIDIA NeMo systems. Presenting it solely in Python would have posed significant challenges during deployment, as it would require adjusting the CUDA version on every machine, rewriting the kernel, and making it nearly impossible for general users to test the demo.
+
+For this reason, I chose to implement it in JavaScript to facilitate deployment on Vercel. However, Vercel's virtual machine does not allow Python execution, which forced me to use NVIDIA's APIs to make some components work. This approach allowed me to create an interactive demo that was accessible to the general public. Below, I provide the links to the demos, which you can easily explore:
+
+DEMO 1
+DEMO 2
+DEMO 2D Model
+Lastly, I hope you can understand my perspective. In my country, an RTX 6000 is valued at approximately 8,000 euros, and my current computer is second-hand, purchased for 300 euros. For this reason, although I never sought to win, it was inevitable for me to feel excited about the possibility of recognition at this level.
+
+From the bottom of my heart, I hope we can move past this incident and maintain a friendly relationship based on our shared interest in technology and innovation.
+
+Warm regards,
+Francisco Angulo de Lafuente