Machine Learning R&D

I have figured out a way of doing machine learning right. Turns out that all popular architectures are just instances of a more general approach. Infinite transformer window is possible. Attention is just a particular case of Hopf convolution. Verification is possible. Faster training, faster inference. Possibly interpretability. But it goes beyond transformers. The machine learning process is analogous to renormalization from QFT which in retrospect makes sense. I have recently written a paper on this however a lot has changed since then. I'm in the process of implementing a framework that implements these ideas. I'm looking for ML R&D position with a company that appreciates this approach. Contact me to learn more. I have a discord channel.