This is an abstraction based article and hence you may want to skip it if you are a only-tools person.
We all love to get into that state. The flow state that is. When the circuit finally completes and ideas are precipitated to tangible models and outcomes. It is like intuitively you understood how things would culminate but they form the beautiful and meaningful picture only after the nurture of time force.
One such culmination happened when we had discussed the newron highway we were going to create in the form of newron.ai plumbing ecosystem helping delivering the mission of the re-electrification of the world through AI.
I do not know the correct etymological origins of the word "neuron" but I shall hazard a guess and say it has to come from the same understanding of a fundamental particle in a system as did the word for an "electron". So we know a neuron constitutes a fundamental element in the realm of the brains and all things associated with cognition and thinking. The whole idea of a neural network that is developed in and as Ai models is nothing but an extention of a similar paradigm in principle. I am no ML scientist and hence won't proclaim an exactness in my discussion which may be technically off here and there. But we do not need to be technically correct here. But more try to make AI a force to reckon with in our daily lives. That is the promise I reckon? Just like electricity is a force of nature harnessed through the material enhancements.
So if my basic physics serves me well, an electronic in motion constitutes the electricity we work with in our daily lives. A static electron is not of much use to us that way. What however works wonders is that voltage difference and that presence of the grid. And "electron in motion" creates all the difference. Just like a neuron in motion fired in our brains I guess? But that elemental velocity is not achieved yet on the AI grid I reckon. There are many hickkups. The ML engineer develops the model or the neural network through which he ought to pass rhe dataset but the data is not big enough. Or the deployment of the model is a task she is not accustomed to as she is not a devops or software engineer. The engineer depends on someone else to deploy the model into the cloud of choice. And there is no performance tracking of the various reruns based on different weights and biases given to the model. And the business leader is completely disaasociaced with the assigning of these weights and biases as pointed out by one of my friends. All in all, there are so many roadblocks and pits on the highway that the speed of the proverbial neuron is slowed massively. And not just within the Neural Network, that is. Not just within the various layers of the NN through which the neuron does its forward march and the backpropagation. We need to understand the velocity of the neuron that matters is the "Business Application" velocity and not just inside the model. We thus need to enhance the velocity of delivery of the proverbial neuron on the highway of our AI delivery pipeline.
When an electron or a bunch find that unhindered highway or a good conductor and grid of transformers and lines, the electricity transformed the world. We need the proper AI conductor highway to deliver a similar "neuricity" that shall transform the world in the decades to come ahead. Only then shall the world witness an AI based re-electrification when we achieve that neuricity in abundance that is ubiquitous and impacts every system. And that is the conductor we build at newron. The one copper wire that delivers the "AI" to the world or the "neuricity" - a term I like to call AI by personally. For the rate of neuron firing inside of your brains constitutes the neuricity of your brain or the intelligence in your brain.