We are going through a gold rush. Companies have started investing billions of dollars in AI projects and there is one career role that is going to benefit the most out of this boom. AI Engineer or ML Engineer. In my company, Atlik Technologies, I hire AI Engineers. I have previously worked with Bloomberg and NVIDIA. Based on my industry experience, I am going to discuss A roadmap with week by week study plan using free learning resources and checklist that you can use to become an AI engineer. AI engineers make highest amount of salary among all the technical roles. What I am showing you on the screen is just a range. The exact salary depends on your skills and experience, the company that is hiring and the location. Now obviously if company is paying you such a high salary, they will have a high expectation. Therefore, Preparing for AI Engineer requires a lot of hard work. If you're looking for any shortcut, then please leave this video right now because this roadmap will require 4 hours of dedicated study for 8 months and that will help you set a strong base. The actual learning is a lifelong process. Before you begin the study, you need to figure out if this is the right career for you. And the way you can find it out is you need to evaluate If you have interest in coding and math, AI job requires strong coding and math skills. And without that, you cannot become an engineer. So if you don't have interest or skills in any of these two, then don't go for an engineer. This is not the end of the road because there are other career roles as well, such as AI sales representative, AI product manager, AI ethics executive, etc. We will not only talk about tool skills, we will also discuss core skills and all learning resources associated with it. Once we have discussed upskilling, we'll also discuss how you can showcase your work to the world so that you can get an interview call and crack an interview. In case you are confused about data scientist and AI engineer role, think of it as data scientist plus software engineer is equal to AI engineer. That's a very simple way of looking at it. Here is a roadmap pdf you can download it from video description below. Requires eight months of study, four hours every day and The week zero starts with proper research folks. There are so many scams going on in the market. So if you buy a wrong course or if you learn from an instructor who doesn't have industry experience, let's say, or who is not legitimate, then you will get into trouble. Nowadays, you see many YouTubers, many people teaching on online learning platforms. They claim that their courses are the best. But when you look at their background, they don't have experience. how to talk nicely and they conduct all kind of scams we have created a couple of linkedin posts which you can look at it and make sure you don't get into those scams we are in fact running a scam awareness program once you have done enough research week one and two will go into learning computer science fundamentals if you have computer science degree you are covered you But let's say you don't have computer science background, then I will suggest you go through this Khan Academy course which covers the basics such as bits and bytes, storing text and numbers, basics of computer networks, HTTP, World Wide Web, basics of programming and so on. Once again look at this particular equation. You need to have solid software engineering fundamentals in order to become AI engineer. Here is the course. which is free and you need to just finish the first four modules remaining modules you can go over it if you have interest in time but the four modules are good enough Khan Academy is a very good platform this person teaches you very well along with the practice exercises in week three and four you will focus on Python Python is the most popular programming language in the AI world today learning Python is actually very easy You need to start with these basic concepts. OK, and we have a playlist on YouTube. This is on my channel and the other playlist is on Corey Schaffer's channel. You can refer to whichever tutorials you feel comfortable with. In this stage, I would suggest you go through only first 16 tutorials because that will cover the beginner's logic in Python. And as an assignment. you will have to finish all the exercises. So if you click on this link, you will see all the exercises, okay? So let's say there is an exercise and there is a solution link as well. I know you all are sincere students, so you will practice on your own and then only you will look at the exercises. Now, in this time period, week three and four, Along with Python, you need to learn some soft skills. Which are those soft skills? Well, you need to build a LinkedIn profile. LinkedIn is the platform that will help you get a job eventually. Therefore, you should not wait until you're done with all your technical skills. The approach I suggest is in parallel, you will start building LinkedIn profile and all other softwares. Okay, we have created a checklist that will help you make your LinkedIn profile stronger. All you have to do is follow all this guideline, check, check, check. And once you have checked all these checkboxes, your LinkedIn profile will look nice. Once you have covered the Python basics, week five and six, you should focus on data structures and algorithms. ML engineer or AI engineer, you'll be writing programs which needs to scale. So you need to know the trade-off between memory and CPU. You need to have deeper understanding on how the data structures work underneath. And for that, we have once again a YouTube playlist. It's a free learning resource. The playlist contains exercises as well. So you can go through. all the data structures and algorithm it will take you two weeks time period to go through all of these and also practice those exercises now this is going to be a long learning journey it's very important that you keep yourself motivated for that i have included some inspirational videos for example in this video i interviewed tanul singh who was a mechanical engineer and he used Kaggle platform to become ML engineer. He's sharing a lot of useful tips and insights. So please go through this video while you are learning your technical skills. In week seven and eight, now you can learn advanced Python, such as what is inheritance, generators, iterators. When you are writing big programs for enterprises, which scales well or which is doing a huge volume of data processing, knowing all these concepts are going to be extremely beneficial. When I was at Bloomberg, We are using generators and iterators a lot because we used to deal with huge objects and when you're running a full loop, it's hard to keep those objects in memory. So using generator, you can return it on the fly. List comprehensives are going to be super important for optimizing your program. Multi-threading and multi-processing is very useful when you want to utilize your computer resources such as the cores or even multiple processors within the computer to achieve a high throughput. For this, once again refer to the same playlist but in this you should go through video number 17 to 27 and all of these videos have exercises so make sure you watch the video and cover the exercises in all these videos i will be talking about theory then we'll be writing some code and then there will be an exercise in terms of soft skills you should start following some prominent ai influencers on linkedin one of them is for example nitin he is a head of ai services at google and he writes posts which will talk about the current trends he will also talk about the hiring trends that he's seeing because he himself hire a lot of AR engineers in his team I find reading his post to be extremely useful the other person is Daliana she writes mainly on data science AI and data science are kind of overlapping so therefore you can follow all the posts on data science as well So, follow all these influencers and spend let's say half an hour every day. That way you're keeping yourself up to date and also you are becoming active on LinkedIn. You should also start commenting meaningfully on those posts. When you comment on anybody's post, what happens is your post, if it is having valuable content, your post or your comment, then it will get an engagement. So let's say this post got 155 likes. Some of these people who are giving you likes could be hiring managers or they could be AI engineers working in other company. So this way you are building a relationship with those folks and tomorrow when you're looking for a job, maybe they will give you a referral or maybe they will hire you in their own team. Building relationships on LinkedIn is super important and you posting comments, valuable comments, okay, don't post comments like this, okay, true, absolutely right, because that will not generate any engagement. You're not adding any value. But when you add some value to the post, you are omitting your personality in this online world of LinkedIn and that will help you build a relationship that will help you get attention to your profile. Remember that online presence is a new form of resume. Along with online presence, you need to also think about business fundamentals. As an AI engineer, you will be working in some industry, retail, finance, any industry. If you have good understanding of business concepts, it will help you communicate better with the stakeholders which are involved in your project. To learn the business concepts, I will suggest you follow this Think School YouTube channel. OK, so I'll show you one video where he talked about how Amul beat the competition. And here. He's talking about numbers and strategies and dairy industry in general. So when you go through these kind of business studies, you are building your business understanding. You are developing a business acumen. OK, additionally, you need to learn the art of asking questions. Discord is a platform which allows you to ask questions while you are learning, let's say Python, SQL, whatever. And if you have a question, where should you go? One of the ways is asking questions in. Discord server. Okay. Now there are many Discord servers for code basics for our channel. We have this Discord server, which has around 33,000 members. Okay. And if you have question, let's say for math and statistics or let's say for machine learning. You can post a question and the community members will answer those questions. Now asking questions is an art. Do not just copy paste the error that you are facing and ask for the help because then people will not help you. Because you are looking for spoon feeding. The right approach is to look for direction, not the spoon feeding. I have highlighted that art in this particular LinkedIn post. I have linked it here. You can go through it. And your assignment for this time duration will be to write meaningful comments on at least 10 AI-related LinkedIn posts and note down your key learnings from 3 case studies on Think School and share them with your friend. As and when you are finishing those assignments, you can just keep on marking them. That way you are tracking a progress. As an AI engineer, you will not be working alone on a project. You will be working with a team. Now, how do you collaborate with the team? How do you share your code with the team? How do you review that code? Well, the way to do that is via version control. Therefore you need to have sound understanding of version control systems such as Git. GitHub is a website which is using Git as an underlying version control system. There is another website called GitLab too. And GitHub is very popular. Develop an understanding of how Git and GitHub works. The topics you will learn are listed here. In terms of learning resources, once again, you can use YouTube. On YouTube, you can refer to Corey Safar's playlist or I also have a playlist here. In this playlist, I have explained things as if you are a high school student in a very simple language using a practical approach. Keep the motivation high. I have linked an interview. of a mechanical engineer who became deep learning engineer using self-study. Mahadev is the name of the person and I love his confidence and the way he approached his entire journey is really inspirational. So I would highly recommend you watch this interview. When it comes to soft skills, presentation is the most underrated skill I would say. For this, I would suggest you watch this Death by PowerPoint video. This video is a goldmine. It is giving you very simple and very powerful tips of how do you build effective presentations. As an AI engineer, you will be working with stakeholders, you will be in a meeting rooms, you will be presenting all the time. And if you don't know how to present well, there is no use of your technical work because you're not able to sell your work or you're not able to convey your ideas in a language that the business stakeholders understand. Watching this video and preparing skills for presentation is going to boost your career. Week 10 and 11, we need to focus on SQL and relational databases. As an AI engineer, we will need data to train our models and to do variety of operations. This data is often stored in a relational database. SQL is called Structured Query Language. It's a language that you use to query data from those databases. Here, you need to learn all these topics. In terms of free learning resources, We have an excellent Khan Academy SQL course, so you can go through it, learn those skills. You can also use W3Schools or a platform like SQL Bolt, which allows you to practice SQL while you're learning it. So I really love this platform. You should definitely try it out. And then on YouTube also, there are tons of video. My channel have this particular video which goes through SQL skills. There are so many other high quality. SQL tutorials available on YouTube in case you want to speed up your learning and you want to learn in a very practical approach and also work on an industry project then I have this SQL course okay this SQL course is very highly rated it's very affordable and we are not only going through all the SQL technical fundamentals but we are teaching how these SQL projects are executed in the industry so all the stakeholder management skills project management skills are also covered for assignment you need to work on SQL resume project challenge On our platform codebasics.io we run this free resume project challenges where we share problem statement and data with folks and people work on these projects and not only that they build presentation and they present it on linkedin so let me show you so here is the resume project challenge where you see the data set the mock-ups everything the problem statement so many people participate in this one and the winner for example here is Arun Sharma so if you click on this LinkedIn post what he did is he built a solution in sequel and then he created a link green post where he explained the solution that he built not only that he attached a video presentation where he was talking as if he's presenting this to business stakeholders now when you are doing this kind of activity you are showcasing your verbal, your written English communication skills to the world. Let's say if a potential hiring manager watches this video. They will get a lot of clues about Aryan's personality, his technical as well as his soft skills. The fact here is that Aryan literally got hired in a company as a data analyst just based on this particular resume project challenge. So this is really effective. It has worked for Aryan and many other folks and it can work for you as well. Next comes Nampa and Pandas and I have attached the playlist and learning resources for it. numpy and pandas are used for data cleaning data exploration those kind of things so you're spending just one week in learning this basic libraries and later on there will be a time period here where you will actually practice the eda skills exploratory data analysis skills then comes the heavy module math and statistics for ai math and states is the foundation for ai any ai project so if you're working as an ai engineer You need to have sound fundamentals in math and statistics. Now, math and statistics is a vast field. I have listed down all the topics which are needed by an AI engineer. OK, so just focus on all these topics. I have also linked the learning resources, which includes Khan Academy's course, the YouTube channels, you know, channels such as StatQuest. There is a free YouTube playlist. And. A channel called 3Blue1Brown. This person teaches mathematics in a very visual and very appealing way. So just refer to his videos if you are interested in learning things like calculus, linear algebra, etc. I also linked my math and statistics course here, which covers all these fundamentals. It also covers an industry project where we had a database of half a million records and data. we did hypothesis testing on the launch of a new credit card okay so you can refer to this course if you want to learn using industry style project based learning next one is exploratory data analysis you might have heard this term eda eda is nothing but you get all the data that you need for your ai project you need to first do some exploration there might be a lot of bad values you need to clean those bad values You also need to perform certain data transformation. Okay, so this module covers that. The technical skills that you need for this are NumPy, Pandas, Matplotlib, etc, which you have learned previously, correct? But in this particular module, what I want you to do is go to kegel.com. Kegel is a website which is hosting our datasets and competitions related to AI. Here, you will find a lot of useful datasets, and also, the problem statements. So you have to go through some of these problem statements. Okay. And Practice. You will see solutions from other folks as well, but I want you to practice things on your own first and then look at the solution from other people. So the exercise here will be initially during learning, you do EDA using three datasets and then you work on additional two datasets and perform exploration. Now comes probably the most important module. Machine learning here you will be spending week 18 to 21 entire month. Machine learning is a vast field and this particular segment covers only the statistical machine learning. So you need to first cover pre-processing techniques and then model building techniques. The great news here is that we have a YouTube playlist. This is a playlist on my own channel. It has received more than 2 million views. I have explained the theory in a very intuitive way. Then there is code and then there is exercise. So go through this playlist first 21 videos only. When you get a job as an AI engineer, you will be using some kind of project management tool. In the industry right now, Scrum and Kanban are the two popular agile project management techniques. It will be good to have some understanding of Scrum. and kanban i have linked excellent free resources for both of it it won't take you much time so please go through them and here is the assignment you need to complete all the exercises in the ml playlist work on two kegel ml notebooks write two linkedin posts on whatever you have learned in ml on linkedin let's say if you have learned about classification you know let's say logistic regression and if you have worked on small problem statement you can write a nice summary of what you have learned and that will generate some engagement so being active on LinkedIn is going to be a constant requirement in week 22 we will be looking at MLOps MLOps is similar to DevOps if you are aware about software engineering in software development there is this role called DevOps where a person will look into You know automating some parts of a software development. So they will be working on CICD pipelines, on Jenkins, on automating workflows, integrating lintas and many other useful tools in GitHub etc. Similar to that MLOps is a field where you are trying to automate some of the things in machine learning project development. Here you need to learn what is API. And then FastAPI. FastAPI and Flask are the two popular frameworks that people use to write server around a trained model. Once you're a trained model, you will write this server so that it can serve HTTP requests coming from a client. FastAPI is getting popular for which we have once again a free YouTube video which goes through all the fundamentals of FastAPI. and you're creating this sample website and calling fast API from that. Then comes Docker and Kubernetes. These two technical tools are used widely in the industry. Whenever we build any ML solution, we usually put them in container. And Docker is something that helps you with containerization. And you can also use Kubernetes for orchestration. Okay. Also, make yourself aware about at least one cloud platform, AWS or Azure. You don't need to go crazy. Just a fundamental understanding of how cloud works. Create a free account on either Azure. Or AWS, if you're talking about AWS, there is something called Amazon SageMaker. That's a platform that allows you to do machine learning on the cloud. Okay, so on the SageMaker, create a platform, try to run some notebook on SageMaker. MLOps itself is a vast topic and many companies have a separate MLOps engineer role. But as an AI engineer, at least you need to have some understanding of MLOps. So don't go crazy here, okay? Because for detail, there is MLOps engineer. It's a separate career role. But as an AI engineer, sometimes when you are working in a small company where there is no separate MLOps role, you will have to do some of the MLOps. All right. So just having fundamentals clear is going to be super important. Now that you have learned essential skills in week 23, 24. You will be building some machine learning projects. So I have linked two projects, one for regression, one for classification. Both of these are YouTube playlist end to end projects, including deployment. Please go through them. In terms of soft skills, you need to build an ATS resume. Don't build resume towards the end. You can start building resume right now. ATS stands for application tracking system which many companies are using and they will use this system to filter out your resume so make sure your resume is ATS compliant so that it doesn't get filtered automatically by ATS system we have created a video on this topic so please go through that video and there is also a checklist that will help you make your resume ATS compliant so just go through all this point check check check and once you have checked all the boxes your resume will indeed be ATS friendly other than resume you need to build a project portfolio website we have linked some resources here so for example I'm going to show you one sample a project portfolio website this website is like your own website where you are writing about your skills what kind of projects you have worked on and you will give a link to a github or whatever that online tool is where you are showcasing your work and here are some ideas for the assignment the projects that we have done on youtube maybe you can start using different technology for example instead of flask use fast api okay in classification project instead of sports celebrity classification you can use classification of movie stars or maybe your family member pictures you That will give your project a unique flavor and it doesn't look like you're just copying a project from YouTube. Now comes a very hot topic, deep learning. You will spend three weeks learning about what is neural network, the fundamentals of convolutional neural network, sequence models such as RNN, etc. Deep learning is getting very popular. It is the biz of GenAI, LLM, ChatGPT. All the hype that you're seeing is using deep learning underneath. For learning deep learning, there are two playlists I will refer you to. So the TensorFlow is a framework from Google. We have this very popular playlist on YouTube. Once again, exercises, code theory, everything is covered. So, let's get started folks. All the learning resources are available for free. All you need is a willpower, motivation, a computer and a stable internet. And then comes end to end deep learning project for potato disease classification. In this project, we built a mobile app which any farmer can use to take a picture of a potato plant and it will tell you whether the plant has a disease or not. Underneath it is using deep learning and convolutional neural network. Week 28 2.30 you can either learn NLP or computer vision you don't need to learn both there will be AI engineers who will be specializing either in computer vision or NLP it's like you become a general doctor and then you become lung doctor or heart doctor you don't need to become both in terms of NLP these are the topics that you can learn there is once again a youtube playlist that you can use to learn theory practice coding and also work on exercises. The last two weeks of this entire eight month long journey will go in learning LLM and Langchain. These are the buzzwords and Langchain is a framework that is getting very popular and if you look at any machine learning engineer positions nowadays majority of them require you to have some exposure to Langchain framework. So for this also, I have a playlist where we have covered all the Langchain fundamentals and we have built three projects, three LLM projects, which you can use to learn as well as you can put those projects on your resume, obviously with some customizations. Remember that in this eight months you have learned all the fundamental skills, but that doesn't mean you have become an expert AI engineer. The learning for AI is continuous. So many things are happening. every day. Therefore, from week 33 onwards, you will be working on more and more projects. You will be working on building online credibility through LinkedIn, Kaggle, and then you will be applying into jobs. And if you have prepared with sincerity, you will definitely get a job because there is a huge boom and there is a lot of demand for people who know AI well. Now I want to share tips for effective learning as well because there is lot of things that you have to learn and you want to make sure that you spend less time and learn effectively there are Some rules for effective learning. For example, you spend less time in consuming tutorials. You spend more time in digesting, implementing and sharing. Nowadays people do reverse. They spend more time in watching videos and for digestion they spend less time. It should be other way around. If you're spending one hour in studies, maybe 20 minutes or 30 minutes you spend in watching the tutorials and remaining time you spend in digesting then you implement you write some code and you share it with your friends group learning is very important when it comes to sharing in our discord server you will see partner and group finder channel where people say okay i want to learn data science uh who wants to partner with me and this way people make groups and then they have weekly zoom calls where they check progress of each other you know it's like a going to gym with bunch of friends if you go alone you will get bored but if you go in group you will stay motivated that's it folks i wish you all the best once again check video description for the pdf the entire pdf is included here all the learning resources are free i wish you all the best if you have any question post in the comment box below if you like this video please share it with your friends we are putting a lot of hard work in making these videos so if you can share it with your friends Or if you like it, it's going to help us a lot. Thanks for watching.