knowing how to use Ai and how it works is crucial now no matter who you are maybe you're just a casual Chad gbt user or maybe you've got your s set on becoming a data analyst either way knowledge is a serious power these days so if you want to learn AI from the groundup this video I will give you a complete road map and show you exactly what you need to do to become your own AI Master we'll talk about educational materials courses programs and all the different ways you can start learning AI so grab
an outad maybe a snack let's Jump Right In the very first step col it Step Zero is asking yourself why do I want to learn AI do you want a job in the industry or do you want just to get better using AI toes because that answer decides whether you're taking the easy route or the hard route for instance if I just want to understand how AI works so I can be more confident using AI TOS I would read online guides tank around constantly and maybe watch a few tutorials eventually I'd get really good at
using AI platforms that would might not bring any big life changes but if I'm learning AI to land a job that's a whole different story I need to master coding fundamentals learn data science dig into machine learning and more the upside is huge though opens doors to a bunch of career paths recently the world's economic Forum published a job report for 2025 and Beyond and in this 290 page document they claim that 92 million jobs will be dis spliced by 2030 on the upside 170 million new jobs will be created and the most relevant skills
that you're going to need to tap into those 170 million is drum roll please Ai and Big Data if we jump on length n right now and search artificial intelligence in the US with a salary filter of over 120k will'll see more than 15,000 results for example there is a lead artificial intelligence engineer raw forbs offering between 200 and 250,000 a year another one is research engine Google from 100K to 200k and we all know Google perks are legendary free food free laundry you name it there's also an AI policy analyst in San Francisco with
a salary of at least 140k and to top it off there is a product manager for AI projects starting at 160k and these AI related jobs are only going to multiply sure AI can already do basic coding so Junior developers are sweating a bit but AI still needs people to manage it to refine it and keep it running smoothly so if you want to join the game now is the perfect time to hop in the train before it leaves the [Music] station now how long does it take to learn AI that's tricky it depends on
how much time you can commit your background in Tech and a whole bunch of other things let's say you're self-learning if you've already done some coding picking up the AI fundamentals could take few weeks or maybe a month if your only AI experience is typ in prompts into Chad gbt you might need a few months to really get the hang of it and if all you want is to write better prompts you can probably learn that in a day check out our guides on Chad gbt Gemini M Journey sore and so on to go beyond
that and develop the advanced skill you're looking at around a year assuming you're not studying all day every day if you dedicate more time you can speed it up the biggest perk of self-learning is that you can directly Target the skill your dream job requires skipping stuff you don't really need because the AI industry grows faster than you can blank new TOs new models and Technologies pop up daily and to stay relevant you have to be on The Cutting Edge the whole time that's why we created GE Academy we'll give you new TOs education materials
news about Ai and most importantly Community all carefully curated by experts with regular Updates this is a nice and convenient way to learn universities on the other hand will teach you everything under the AI umbrella which is awesome but also means 3 to four years fullon studying so Peg the PAAD that fits your goals in the end AI isn't going anywhere it's important to understand that AI is a general term that includes a bunch of different fields a lot of people think AI is some single Mega technology that can do everything in reality with good
things like Ani artificial narrow intelligence all those everyday TOS we use without even realizing it then there's AI artificial general intelligence which is supposed to be pretty close to how humans think and someday we might even get ASI artificial super intelligence which would surpass our own intelligence that's basically Skynet territory but that's not what you're going to study because when you look at AI from your perspective the hierarchies is like this at the very top is AI the broadest field all about Building Systems that can mimic humans reasoning solving problems making decisions all that brainy
stuff inside AI we have machine learning ml is a subset of AI focused on teaching computers to learn learn patterns from data so we don't have to code every tiny instructions give it enough data and it figures out the rules by itself that's how Netflix picks the next show you'll bench based on your watch History it's always to learn about you whether you like it or not now deep learning is a subset of machine learning it's basically inspired by how our brains work using layers and neural networks to handle super complex data that's what powers
facial recognition or even TOS like Chad gbt at the bottom of it all the most corner own discipline is data science data scientists might tap into AI ml or just basic statistics depending on what problem needs fixing and yeah let's be real you probably won't learn data science overnight unless you dive into some serious study this is the kind of thing universities excel at so that's my personal recommendation but you do you choosing between deep learning machine learning and general AI isn't exactly straightforward unless you actually try them out think of it like a letter
start with General AI ideas move on to ml then if you feel confident jump to deep learning and maybe if you're feeling extra ambitious try data science after you figure out which area excites you most you have to pick an approach coding or no coding the names speak for themselves no code is simpler but more limiting say you want to generate text if you can code you're basically stuck with Chad gbt gemini or any third party models developers have put out there you can still do some cool stuff like train those models on your own
data or tweak certain parameters usually with zero coding but with that you're locked into whatever those Services let you do if you suddenly think hey I want to fine-tune this model in a totally new way you might not be able to that's where coding comes in it's tougher sure but the payoff is huge instead of being stuck with whatever's on offer you can take existing models and customize them change the internals or even create brand new models from scratch think of it this way if you can code you're basically the person who catches the fish
if you go no code you're the one waiting for someone else to hand you a fish Playa that's the difference in control and potential for a coding approach your best first step is python it's probably the most beginner friendly language out there that syntax is clean and easy to read and you can run python on virtually any OS even in your web browser on top of that the community is huge so you'll never struggle to find tutorials or answers to your questions may not be CP in certain areas but for new developers python is perfect
also has a giant Library ecosystem including popular Frameworks like tensor flow or pytorch for machine learning which can save you a ton of time when building or training models and it's free so can start coding right away aside from just learning how to write python code it's crucial to learn GitHub think of GitHub as a global meet in place for Developers where they share projects and collaborate almost like GE Academy people share their knowledge and learn from each other on GitHub you can find everything from small code Snippets to massive AI models or entire pipelines
exploring these repositories is a fantastic way to learn you can also showcase your own work there for potential employers your GitHub profile basically becomes your code in resume sure GitHub might seem like just another website but it's actually version control system that helps you keep track of every change you make to your code there are lots of simple lessons on YouTube that will help you get comfortable with making commits creating branches and collaborating through poll requests finally the most important skill you can develop is reading other people's projects you have to peel back the layers
and see how everything is put together that means learning to reverse engineer someone else's project taking it apart bit by bit and then piecing it back together in your own way so go ahead and download a few few AI or ml projects that line up with your interests and push your skill level just a bit then dig into the source code to figure out what each function or class is ding once you've reverse engineered a couple of these you will be amazed at how much more comfortable you feel with AI as a whole and with
your own coding skills and if you are on the uh no code route that same principle applies but you will be studying how other people structure their prompts and refine their outputs instead of analyzing [Music] cod as a starting point I'd suggest learning llms first we all have used them so understanding how they work won't be super difficult start by watching an hourong video by Andre Kathy on LMS it walks you through Transformer models in a way that's surprisingly easy to follow and you will come out with a deeper understanding how llms are built especially
helpful if you ever decide to create your own down the road once you get a handle on what LMS are and how they can be used for chat style tasks it's time to try out various apis open AI is a top Contender with its GPT 40 they've added a lot of features in the past year including an assistant API that behaves like an agent plus a realtime API that can take voice and respond quickly with text or audio that just keep in mind that open AI billing can be confusing if you're going to try that
be sure to first drop a question in the Discord chat so you get a smoother start the pace at which AI evolves is crazy Will Smith iten spaghetti was a benchmark a year ago and now Sora doesn't even blank when getting requests like that the hall industry grows faster than any other industry in the world navigating all this isn't easy has the time to set through all the news and clutter to find stuff that's really worth knowing about our team at GE Academy you will get immediate access to the latest AI news in a chat
where you can meet other AI enthusiasts and talk to our team and for the price of a coffee you can get full access to our platform even more fresh info with AI trans reports that will keep you up to date with everything happening in the world of AI we have a special AI toll finder that will help you find the right toll where your exact needs in a separate tab you will find curated list of the best AI TOS updated weekly all with Comprehensive guides on how to use these TOS you'll also get tons of
promp templates we also encourage everyone to show the results in your AI showcase Channel and will'll be given Awards to the best creators we want to create the biggest and the most useful Community for AI enthusiasts ever so click the link in the description and become a part of G Academy AWS and Google both have their own ecosystems bedrock and vertex AI where you can pick a model fine-tune it deploy it and then hook it up to different TOS or apis Bedrock even offers an agent concept that's useful for building more autonomous AI apps hug
in face is your best friend if you want open source models you can run privately there are tons of models here like metas Lama series which can consistently rank high the big benefit is that you can find tune or even deploy your own version to save cost and keep everything private if you ever need to run models locally look into Llama Or LM Studio both let you download models right to your computer and make them available locally just remember that for anything bigger than 32b parameter model you will need a beefy computer with a really
strong [Music] GPU after you somewhat feel confident with LM try looking into agents if you want to build agents yourself there are several Frameworks in python or JavaScript typescript that use Simple interfaces to make the process simpler Lang chain and Lang gra is popular though it's more complex than it used to be Microsoft's autogen framework originally part of autogen Studio got split into two versions when the founder left which can be confusing crew AI is all about multi-agent collaboration though I personally haven't tested it much open AI SRM is probably the easiest entry point for
a setting up simple agents and flows AWS Bedrock I mentioned before offers UI based agents and workflows with lots of features but you will need to be familiar with AWS Basics first Lama index is another good pick if you want an easy way to work with multiple data sources like single databases that include SQL Json or vector based queries and only after you feel confident with agens and LMS you can move on to rag or retrieval augmented generation it's basically when your model responds to queries without prior training or data to use but for now
let's not dwell on that right now that's to Advan for this video when you're just starting you need to try as many things as possible this is the only way to find out what you're truly interested in what works for you if you feel that image generation is something you want to look deeper into maybe you should also check out the video generation models or if you're into language processing then maybe you should also so take a look at data science so to make your learning A bit easier here are some materials for you one
introducing to data science and python course from Michigan University takes only 34 hours to complete focusing on Python Programming and data manipulations and it's totally free so be sure to roll Microsoft AI classroom is also good starting point walks you through the fundamentals what AI is what it can do and ties everything back to Microsoft Azure cloud services it's not a super long course so you can finish in just a couple evenings if you are motivated plus it's completely free you'll learn how to deploy simple AI projects in the cloud and get a taste of
the real world scenarios Azure can handle IBM also has a free course that covers core AI Concepts ethical considerations and everyday applications giving you a thorough overview what AI can do businesses and Beyond it's well structured so you won't get loss and jargon short useful just what you need right now now these basically courses are definitely useful no doubt about that but personally I've always felt a little weird studying just from courses being in an AI Family Sharing knowledge that's what makes a real difference and that's what geek Academy is all about we are building
the world's best and biggest AI Community to bring AI to millions of people join to stay updated in the fast world of AI and get ahead of the curve I would also suggest checking out the elements of AI by the University of heleni this one focuses not only in technical aspects of AI but also shows how AI is affecting Society jobs and pretty much every industry it's very beginner friendly and completely free if you are curious about the ethical and social dimensions of AI like how chatbots and data analytics might change our daily lives this
is where you'll get those answers in a fun digestible format if you like learning by actually doing stuff machine learning crash ques by Google it's a great pick it's full of interactive coding exercises that show you ml Essentials like linear Reg classification and even basics of neural networks you will use Google's teaching tools to experiment with real data sets which is awesome if you're itching to write some lines of code and see immediate [Music] results when you finish all those and feel confident in your skills move on to something more advanced and specific by now
you already know exactly what branch of AI you want to learn great learnings introduction to AI for beginners is a good option that merges Basics and somewh advanced stuff it starts introducing practical TOS like python early on so you can begin coding basic AI scripts right away it's still free which is amazing considering how covers both conceptual knowledge and Hands-On practice by the time you finish you will know enough to start tackling small AI projects on your own I will leave a link to GitHub page with tons of links to popular courses and education materials
in the description cheers to the AI enthusiasts who created this list to make the learning process smoother and more effective try to jump into real world projects as soon as you can theory is great but you don't want to get stuck in book mode forever right with AI it's all about Hands-On practice that's why in all our video tutorials we push you to follow along our lessons might not be as hardcore as coding from scratch but hey it's still practice and practice is how you get good while you're grinding through these projects don't forget that
Community is everything when people gather around a common passion especially AI can really boost your motivation Community helps you when you're stuck celebrates your wins and share as cool ideas you never would have thought of yourself we are building that kind of community in geek Academy and it's designed so folks can share AI knowledge show off what they've made and basically help each other out we want learning AI to be easy for everyone no overpriced courses needed if you're pumped about Ai and want to dive deeper join our server aside from the community stuff I
really recommend looking at a bunch of job posst tanks to see what skills companies want right now and even if you're not job hunting yet knowing what employers expect helps you shape your learning path trust me it's easy to get lost in the weeds if you don't have a road map everything seems important at first right but going to focus on what actually matters at this point I'd focus on cloud platforms like AWS or Google Cloud for deploying AI models also get comfortable with tensor flow torch and ssit learn these are basically the big three
you see everywhere obviously prompt engineering is another area to keep improving prompt engineer RS are already popping up and it could easily be the next beginner friendly position in AI python is awesome for training models but don't be afraid to explore new stuff take matlb for example you might remember it from Skull but guess what it's also used for developing AI models blew my mind when I first found that out once you are decent at python it's a bit more maybe check out C++ it's a bit more complicated but lets you do tons of coal
powerful things you don't have to master it just learn enough to get a better feel for how AI Works under the hood somebody in our community probably knows it and can back me up on how handy can be also test out different AI TOs and models a good weekly update list of the best AI TOS you can find in geek academ's Discord server with a detailed guide for each toll each of those TOS has its own quirks and Specialties and you never know what might become the next big thing so dabble in Runway ml s
or deep artifacts if you want to generate videos for natural language processing try open AI gbt playground or hug and face Transformers just play around and see what you like and yes start making money from your AI know how as soon as you feel comfortable there are tons of ways to do this from selling prompts to launching AI theme products like t-shirts hoodies if teaching is your thing you can upload your own courses on platforms like teachable or you can jump straight into small business Consulting because trust me there are lots of small companies that
need a little AI help as you get better you can sell specifically train models on sites like model Depot start a YouTube channel about AI or go after an entrylevel AI job there's no shortage of ways to turn even basic skills into a little or a lot of cash the key thing in learning AI is to stay consistent and have a solid plan know exactly what resources you're using and follow a road map so you don't get lost in all the AI hype honestly AI can be easier to learn than other fields because it's so
diverse you can specialize in one type of allll become a developer sell proms or even consult companies it's hard to do it wrong where there's so many ways to succeed especially when you have a community backing you up so if you haven't already hit the link in the description to join us and thanks for watching and we'll catch you the next one
