I’m making the most detailed engineering series about robotics. We will cover mechanical, electrical, and software engineering behind robotics, inference, artificial intelligence, data centers, and Kardashev 2 scale technologies. The Potential Dropout teaches you stuff #thepotentialdropout #mechanicalengineering #electricalengineering #artificialintelligence #robotics

Within the next 180 seconds, you're going to know a bunch of keywords that my ADHD brain is going to come up with concerning robots and everything you will probably know after you watch this series about robotics that I'm going to make in the next months or years of my life. Robots have the mechanical side of things and it has the software side of things. On a mechanical side of things, there's everything concerning the motors, actuators, mechanical engineering, finite element analysis to make sure that nothing breaks, design and manufacturing, dynamics, center of mass, kinematics, hardware side of control systems, what it entails and everything related to the electronics, also just the electronics itself and voltage so that stuff doesn't get fried and okay, on the software side, you got everything making this robot smart or the software side of the control system. So what makes a robot smart these days, you got the basic kind of computation going in, but you also got now the frontier stuff going on for robotics. Robots now have AI models inside it, that's why there's a conversation around physical AI. AI for the physical world, so mostly for robots and drones, whatever physical object you want to reach a certain level of intelligence, it could be a table that becomes autonomous and smart and knows what's going on around it and thanks, but anything physical, that's physical AI. Robots, big as example of physical AI, robots run an AI model and it's probably running something called a vision language action model. VLA's, there's going to be a whole conversation about VLA's and what it means to train a VLA model for robots, how to get this data for the training, all of this, it's all going to get covered. And then the really cool fun thing is now that you got the model, how does the robot use it? Well, there's something called inference, so in a very non-technical way, I could say that inference is just a machine, don't the thinking. What does it mean for a machine to do the thinking? That's when we go into the silicon chips, the GPUs, compute, edge inference, cloud inference, does the thinking happen within the robot, or does this thinking happen somewhere else? Let's say in a data center, then there's a whole conversation about data centers and the thermodynamics of data centers and sustainability, then there's going to be a whole conversation about moving data centers, probably elsewhere, so it doesn't melt the earth. That's when we talk about space stuff and Kardashev 2 stuff, and I don't know, Dyson Swarms. But at the end of the day, it's all just rocks becoming sentient and doing linear algebra. It's just math, a linear algebra.