Transcription
Hello all, my name is Krishna Nayak and welcome to my YouTube channel.
So guys, today in this particular video, we are going to go ahead and see the entire road map to learn AI in 2025 and here I'm going to provide you all the free resources, videos and materials. So please make sure that you watch this video till the end.
And why I am actually making this specific video? The reason is very simple because right now generative AI, agentic AI that is the talk of the town. Everybody should at least have the idea about generative AI application, generative agentic AI applications, how you should go ahead and build. Whether you are coming from a developer background, whether you are program managers, whether you are in the leadership position, everybody should know this. The reason is very simple because here you'll be able to derive some amazing projects. Many companies more than 80 percentage of companies more than 70 percentage of companies are working on different different kind of generative AI applications they're building rag applications they're using different kind of LLM models but still the question rises kish I'm a fresher please tell me a specific path kish I'm an experienced professional I'm working in this particular technology I really want to make a transition into AI so for that what should be the kind of road map there was also question like kish I am 10 plus years of experience I'm not that interested in coding but definitely I want to get into AI right so you tell me a specific pro road map how we should go ahead and learn so in this particular video I am going to cover all these questions in a much more efficient manner this video will be little bit longer because here I've created a very efficient road map within 8 months of time frame you should be able to complete this if you are starting from scratch right other than this just follow this particular road map I have provided free videos resources just go ahead and watch it you will definitely be able to do amazing things over here so quickly I will go ahead and share my screen so here you can see that I have written this amazing road map to learn AI in 2025 and here I have you know drawn three different routes one is the traditional route The second one is the modern route and the third one is something called as an advanced route.
Now why do we actually require this routes? I will discuss about this in a much more detailed manner. So let's say uh as I said this road map actually incorporates for everyone whether you are a fresher whether you are an experienced professional whether you are a leader you are in a leadership position and whether you are a person who is also coming from a complete non-technical background. Okay, nontechnical background basically means let's say that you may be working in HR, you may be working in finance and you also specifically want to learn how to get into AI, right? So I will also be talking about them.
So in this actual route here you can see this is my first path traditional route, second part is modern route and the third part is advanced route. So let me go ahead and fill like if someone was coming to me around 2 years back and they were asking Kish what should be the road map to learn data science I would basically say okay go ahead and first learn data science first. Okay so here I will go ahead and write here you have to learn data science first. Now what you really need to learn in data science. Okay here you need to master some amazing things like what do you need to master? you need to master DS that is data science, ML, CV, NLP. Okay. So here you need to master data science, machine learning, computer vision, NLP, right? And in all the specific topics you should be able to understand that stats and linear algebra is included a part of it. So in short, you're mastering all these technologies like data science, machine learning, CV and LP. And whenever I'm talking about mastering, I'm talking with respect to development of a end-to-end project. Okay. So first of all, you need to learn data science first. If you are asking me 2 years back and today also if any fresher who is in the college who wants to really sc start from scratch must definitely follow this road map itself. Okay. Then after mastering this the next stage is you go ahead and master Genai. So here we are going to go ahead and add Gen AI. Now when I say add Genai that basically means you go ahead and explore different different frameworks. Think about like how different different LLM models work. How does LLM training happen? You know how to go ahead and develop a generative AI applications. How to work with multimodel applications. Each and everything comes in generative AI. So all this detailed uh road map I will discuss okay as we go ahead. But right now let's cover this specific diagram this flowchart. Then after you've complete generative AI then we go to the next step wherein you can start learning agentic AI right agentic AI and MCP right now here when we are talking about agentic AI and MCP I'm actually talking about developing agentic AI applications agents you can go ahead and develop rag applications each and everything right so this should be the road map if someone was asking me 2 years back and if you are a fresher and experienced professionals also let's say that an experienced professional wants to start from uh like let's say this person is 0 to 3 years of experience or 0 to 5 years of experience I would basically suggest this traditional route 2 years back okay but now the scenario has changed but even a fresher if if you're calling from college if you're coming from college if you're still in college you have just passed out you want to go get into AI still I would suggest follow this traditional route. This is the best traditional route wherein you're mastering everything like data science, machine learning, CV and LP. This will definitely be your base and on top of that you are adding generative AI and agentic AI skills. Okay. Now when I'm also talking about generative AI and agent AI skills, I'm also talking about developing end-to-end projects, doing the deployment in cloud using some amazing LLM ops tools, each and everything. Okay.
Now coming to the second part which is the modern route right nowadays if many of the people ask me kish which path should I probably go ahead and take in order to learn AI nowadays I usually prefer them modern route. Now what is modern route? Here I will say them hey go ahead and learn generative AI first. Okay. So here you go ahead and learn about generative AI first. In this generative AI you specifically understand about LLMs because at the end of the day it's all about content generation right specifically if you're working with NLP and all right and when I say generative AI first here obviously some of the knowledge of NLP is definitely required okay NLP machine learning it is required okay deep learning some of the uh topics with respect to that is required okay now after you you start with generative AI first of all you master generative AI Master generative AI. Okay. So this is the first step. So once you're starting generative a master it try to understand how to build generative AI application rag applications how to use different different frameworks like lang chain lang graph crew AI all the things and also do the deployment. Okay. On top of this once you complete this go ahead and start adding on top of it agentic AI also agentic AI and MCP let's say okay so on top of this you will go ahead and add aentic AI and then finally once you are covered this right and till you master this right you will definitely be able to get any kind of jobs in the market because if you know the entire life cycle if you have developed some amazing end-to-end projects and here if you have also included LLM ops then you're good to go and work in any kind of industries. Now once you have completed this you're already working it now on top of that you can start learning your DS fundamentals. Okay start learning your DS fundamentals. So this is the road map I usually tell to the people who right now ask me and who are specifically these two type of peoples right freshers and experience and even nontechnical people I've seen many many nontechnical people also following this specific road map because they're saying that hey krish I really want to get into AI I also want to start coding right so all these things will basically be following this modern route so it is up to you whether you're in experience whether you are 0 to 5 years of experience I would still say hey go ahead and follow this particular pattern right because understand to learn DSML CV NLP it will definitely take time if I say about this it will actually take around 4 months if I talk about geni it will take around 2 months only if you are devoting around 2 hours daily right 2 hours daily so you will definitely be able to cover this right here also for agentic you may again take 2 months now rag and all will come either in genative aentic okay Now I hope you are able to understand in the modern route because we cannot you know right now companies are looking for professionals who can work in generative building generative AI application and genetic AI application. So in this specific route I'm saying start learning generative AI first you need to have some basic knowledge about NLP right and then you're good to go and start with mastering generative AI. So once you are able to develop because here you're working with LLM models you're focusing on different different frameworks wherein you're focusing on understanding how to build those kind of applications. Yes, the basic knowledge of NLP, machine learning, deep learning can be asked in interviews. But nowadays still in this modern route, right? If you go for the interviews, they're focusing on building generative and agentic application, how to make rag, how to perform fine-tuning each and everything, right? So they're focusing more of this specific questions. Okay.
Now coming to the third route. Okay. Which is the advanced route who are really really efficient people. You know they can start all three parallel. All three parallel all three parallel. Now these are those people who are already in the technical domain and they can quickly learn things they can start this three right but again the focus must be given in this modern route more right whenever you go ahead get time go ahead and learn the basics of DS machine learning CV and LP and continue this specific thing right and this is nothing but this is basically a comprehensive AI expert so here you actually become a AI expert right so this is what it is basically happening see for leadership ship and nontechnical people what I say usually you can follow this three parallel uh I'll not say for tech nontechnical nontechnical can actually follow this if you really want to get into the coding industry right coding industry then definitely follow this pattern that is traditional route then go to modern route then go to advanced route now you you make the decision you know what is basically required see wherever you learn something let's say if you are going to this modern route some amount of knowledge is will be required from the traditional route also some basic knowledge like NLP, machine learning, deep learning, you know, that will actually set up your base. But it is not like you will be stopping over here. If you start with this modern route, you'll get blocked somewhere. No, here focus more on building applications. Now I hope you are able to understand from this particular road map how you should basically go ahead and do this. If I will say if you are a leader, if you are a very experienced person just go ahead and follow this modern route or either you can also follow this advanced route where you can start things parallelly right but for freshers for the person who have just started want to get into the AI industry I would still suggest follow this traditional route it will take time but understand as you keep on learning things you'll understand the base of each and everything I have spent around 6 to 7 years making videos and I've covered almost everything in my YouTube channel. Just imagine 2K plus videos, right? 2Ks plus video. How many people have actually got benefited out of it? Okay, now this is fine, right? So, as I said, this is my main three stage. This is my first stage, second stage and third stage. Now for every stage I have created a specific road map along with free videos and materials which I will show you as we go ahead. Okay.
Now let me go ahead and open my road map. So here you can clearly see that this is how my road map looks like. Okay. So very clear detailed road map. Over here you'll be able to see this. This is the AI mastery road map. The three core paths to excellence. Okay. And here you can see minimum Python 3.11. You can also use 3.9 3. So here you have three sections. One is the data science and classical AI. Second is generative AI. Third is agentic AI. If you see path wise data science, NLP, computer vision four months I'm taking. If you spend two to three hours here, you'll be able to definitely become data scientist, ML engineer, data analyst. Part two is generative LMS. two months generative AI engineer, AI product developer, agent AI and autonomous system, two months a architect, agent developer and main thing. This is the same route that I have actually explained you right now. Okay. Now with respect to this, this is my section one. I have created a perfect road map to learn everything with the free video materials and everything itself. Right? So here you'll be able to see this comprehensive road map covers Python programming language stats, machine learning, this all you have to do is that just go ahead and click this. Okay? And if you remember this I have already created but I have updated a lot. So here you can see work of a data scientist data analyst. So first of all you need to start with Python programming language. All the Python programming languages basically over here. Then you go ahead with statistics. If you start taking the traditional route in statistics, I have created very detailed videos u in ML 43 videos are just there in statistics very detailed in my YouTube channel at any point of time and today also the highest number of views that I get from is from this specific videos then you have EDA then you have feature engineering then you have uh databases you can go ahead and learn about databases like MongoDB my SQL and all now see I'm breaking that everything right on the high,000 ft overview I have told about you have to learn this but in order to master all those things you require all this important techniques right then you have this machine learning then deep learning then NLP playlist then important framework of production deployment MLOps is the must over here in order to complete data science you if you really want to master you definitely need to know about machine learning operations that is design model deployment and operation so here you have all the free links along with the YouTube video so here if you just go ahead and click it You should be able to see videos. Yes, some amount of ads will be visible but it's okay. Okay. Uh it is a YouTube video so dedicated three three hour four hours videos are available over here completely in depth. Okay. Then you have circle CI then you have deployment of things evidently AI graphana airflow bento ML AWS sagemaker DVC dockers everything is basically there how to develop an end-to-end projects with MLOps and deployment with open source tools. Here also we have created both in Hindi and both in English. Right? If you want to follow that also you can do machine learning, deep learning and LP projects. I have actually listed down around 10 to 15 different kind of projects which you can just go ahead and explore it. Always remember one very important thing is that you need to start using chat GPT cloud or Google Germany Grock efficiently. U the reason is that you don't need to create problem statement for you. You can use this platforms to develop modules quickly get the code become much more efficient become increase the AI productivity because once you're specifically coding you don't need to keep on searching for multiple resources just be at one point think of a specific use case ask it how to probably do it get the description and start building it always take help of AI don't do the development completely from scratch that would be my suggestion okay then here I've actually mentioned each and everything with respect to this okay now this is the first path to master data science in 2025.
Then we go ahead with generative AI. Okay. Now in generative AI, I've created another road map which is my second section and this is for the modern route. Okay. So here you'll be learning about large language model, prompt, diffusion model, fine-tuning techniques, multi model, AI system production deployment. If you go ahead and click this here you should be able to see the entire road map. Okay. So here to get into generative way how you should start what should be the basic things that you should know you need to know Python programming language you need to know basic machine learning and NLP. So here I've developed some amazing five live sessions. These are like big big sessions which you can actually just go ahead and cover it and automatically you should be able to quickly cover all the specific topics. Then basic learning deep learning concepts from day one to day five. Here also I've created amazing videos on this. These all are on my live sessions which you can quickly learn and to get the basic knowledge right. If you want to learn AI, you know, I can also take four years and teach you same concepts, right? But how much is actually required to get into industry all those specific videos is it's more than sufficient. If you tell me to create teach uh neural networks, I can take 2 days session also and still that topic will not get completed. And sometimes I may only like if you if I just tell you that I can also teach you in a way in just 2 hours so that you understand each and everything that is required in order to build a specific product. So that is the thing that you really need to understand. Do you really want to do a PhD or do you really want to learn concepts and start building products that is the most important thing okay now here is the next thing is that you have advanced NLP concepts day six to last video. Okay. So here you'll be able to see that here also detailed live sessions is there already. You can just go ahead and check it out. Okay. Then starting the journey towards generative AI, GPT4, mistral, llama, hugging face all these things. So you have the generative AI tutorials. I have also included with cloud like AWS, Azure, generative AI with Google Germany playlist and fine-tuning. So I have included what are the four main important things that we specifically do in generative AI. We develop generative AI applications. We develop uh we work with generative AI projects in AWS Azure uh for Google ge also I've actually created a separate playlist then here also you have all the finetuning. So with respect to finetuning also you can clearly go ahead and just go ahead and see the YouTube videos where I have actually made each and everything available for you. Okay then you have this vector databases which is also included over here. So in generative AI you should also know about vector store and vector databases and deployment of projects you can do it in AWS Azure use langu and hugging spaces. Okay. So these all things have been completed and I hope you should be able to explore things with the free content to learn the modern route wherein you are trying to learn about generative AI.
Now once you complete generative AI the third thing is about section three which is nothing but agentic AI. Never stop at generative AI because now people are focusing on developing agentic application agents right. So for this there is a separate road map. So in this road map you'll be able to see again how to start with the agentic AI. First of all again you need to know Python programming language some basic knowledge of machine learning natural language processing deep learning for NLP. Then you go over here with end to end project. See I've created generative AI tutorials with project agentic AI tutorials agentic AI playlist with different frameworks multimodel rags each and everything is basically covered over here. And then finally you also go ahead and check out this model context protocol. I have also created a detailed playlist on MCP which will actually help you to understand each and everything. Just imagine guys so much of free content. Uh the best part is that I've contributed so much right that I'm really happy about it. You know people can just go ahead and start exploring things and can get into any industries. If you are just following this then here you can understand best aenti frameworks to learn like lang chain lang graph agno crewi autogen. So for all these things and there is also a live session which we are running in krishnak academy you can go ahead and explore this if you have any question call our counseling team right. So these are the three main section right. So here we have covered section one. This is my section one, section two and section 3. All separately I have created playlist from MLOps to LLM ops to data science, machine learning, CV, NLP. Each and everything is basically there. Okay, you can just go ahead and explore it.
Once you complete this then here is the additional learning resources. See um I'll tell you what is the differences between the free content with respect to the content that is available in the Udemy and live courses also. Okay, as you all know I also sell courses in Udemy in the very affordable cost at 38 3.99. So if you think right you require more help more structured learning then you can just go ahead and follow my Udemy courses. So here I have created another GitHub over here to follow the road map of my Udemy courses. So again based on that track this is much more divided courses like if you want to go ahead and learn mathematics for data science Python with DSA data analyst track data science track generative and engine tracks MLOps and deployment track big data engineering track Figma course track anything that you want you can go ahead and do this and here I in a month I give three coupons on every 5th 15th and 25th. So based on the timeline you can go ahead and use that particular coupon to get the courses in 3.99. Okay.
Now final thing is that these are my additional courses. Let's say you want more handling more support then just go ahead and join our live courses and mentorship. Right. So here you can go into this particular website krishnak.in/live classes. These are all my live classes that is currently going on. Okay. You can join for them. And this is my upcoming AI for leaders and professional. This is specifically for nontechnical people. Okay. And the reason of making this particular course is very simple because I want to cover everyone right whether you are coming from a coding background, non-coding background, anyone. I'm trying to cover each and every people out there with respect to the learning right. So you can go ahead and explore even this specific course.
So this was the entire road map that we are following. How to use your road map? Choose your own path. Start with the section that aligns with your goal. Follow this repository. Each LinkedIn Each linked repository contains detailed week-by-week plans. Complete projects are included in each road map. Track progress. Use the provided checklist to monitor your learning journey. Learning recommendations. Start with section one for strong foundation. Can jump to section two and then you can also explore section three that is agent AI.
Now I hope you like this particular video. My main aim is to cater everyone wherein cost should never be a factor. So I've provided free content. I have even provided paid content. It's all your support. Based on your support only, I'll be able to do more things. But trust me, from the past six to seven years, we we and our entire team has done a really amazing work in making amazing career transition in the field of data science. Wherever we go in any company, they know about they know about me, they know about my team. That is the best thing feeling that uh I I basically give on my work that I do right and that is the best motivation factor for me. So I hope you like this particular video. This was it for my side. I'll see you in the next video. All the information will be given in the description of this particular video. I'll see you in the next video. Bye-bye. Take care.