Getting Started with Claude Code for Absolute Beginners!

Key Takeaways

Today developers are surrounded by AI tools, GPT models, Gemini, local ALMs, copilots and CLIs. But most of them still work the same way. You ask a question, get some code, copy paste, and then spend time fixing and correcting things yourself. That's get tiring very quickly. Cloud code is different. Built by anthropic, cloud code works inside your coding workflow. It understand your entire code base, reasons across multiple files, follows Git workflow and behave like a supervised coding agent, helping you build complete application faster with more confidence. Welcome to this course on clot code, the coding assistant. I am Somil Jan, data science traininee at analytics vidya and I'll be your instructor for this course. I'll guide you step by step from understanding how clot code works to using it directly from the terminal to building real application where AI becomes a true development partner. In this course, you will build games, create intelligent agents, develop web applications, and even design mobile workflows all using cloud code the way it meant to be used. If you want to move beyond AI assisted coding and start building with an AI that can think, plan, and execute alongside you, this course is for you. Let's build with cloud code. Hello and welcome to this course on clot code the coding assistant. In this course we will explore how to use clot code to enhance your development skills and build powerful applications using AI. Let's quickly walk through what we will cover in this course. In the first section we will dive into clot code itself. We will understand what clot code is, its core capabilities and how it differs from traditional AI coding tools. In the second module we will get hands-on. You will build your first agent using clot code and learn how to set it up. We will also go through a mini Super Mario game project and then dive into creating a mock interviewer agent to help you practice interview questions. Moving on to the third module, we will explore how to build a web app using clot code. And we will finish this section with a project where you create a 3D portfolio website showcasing your skills projects in visually engaging way. Finally, in the last module, you will build a mobile lab workflow and we will also work on an AI study assistant to help you prepare for exams by generating tailored questions, MCQs, and summaries from your study material. Thank you for watching this course introduction. Get ready to dive into the world of clot code. We have a lot to cover and by the end of this course, you will be able to use clot code effectively for your projects and task. Let's get started. Hello and welcome back. In this video, I will introduce you to claude code. Understand what it is, why it matters, and how it represents a shift in a way developers works with AI while coding. Claude code is an agentic AI coding assistant developed by Enthropic. What makes it different is how it operates. Instead of living inside a chat window, cloud code works directly from your terminal. It can understand your codebase, edit files, run commands, and assist with real development tasks, making it far more than just a code suggestion tool. AI assisted coding has gone through a clear evolution. Early tools focused mainly on code suggestions helping with autocomplete or small snippets. Today the shift is much bigger. We are moving from suggestion to task level execution and from passive AI to active AI agents. Cloud code represents the shift. Instead of just recommending what to write, it can understand a task, plan the steps, and actively help execute changes across your project, making AI a true development partner rather than just a helper. Traditional AI coding tools have clear limitations. They typically work in isolation, responding to a single prompt. Developers still needs to manually copy paste code into their projects. Most importantly, these tools lack awareness of the broader project context, making them inefficient for complex real world workflows. Cloud code itself has evolved rapidly. It began as a stable CLI based coding assistant and gradually matured into more intelligent developer friendly tool. Today it functioned as agentic workflow system capable of handling multi-step task, maintaining context and supporting automation across real development environments. At its core, cloud code behaves like an AI developer. It can read and understand your entire code base, make save file modifications, execute commands, and reason across multiple steps. This allow it to handle tasks that go beyond simple code generation such as refactoring, debugging and workflow automation. Cloud code is installed locally as a command line interface which gives it direct access to your project files and commands. The reasoning power comes from cloud code models running in the cloud. This combination ensures strong intelligence while keeping developers in control of their local environments. The key difference between cloud code and traditional AI tools lies in the execution and the context. While traditional tools stop at suggestion, cloud code actively participates in development, editing files, running commands, and maintaining awareness of the entire project. That brings us to the end of this video. In the next video, we will move beyond the concept and start using cloud code hands-on to build real world workflows and applications. In this, we are going to talk about the core capabilities of clot code. Now that you know what clot code is, the next obvious question is what it can actually do for you as a developer. That's exactly what we will break down in this session. We will look into the details at different capabilities of cloud code. Let's start with the first capability that is to modify and manage files directly. Cloud code doesn't just suggest what you should write. It can actually create new files, update existing ones, delete files and even restructure your project folder. So instead of copy pasting code manually, cloud code works directly inside your project just like real developer world. Another big strength of cloud code is how well it understand your code base. It can read through the entire project, understand how different files and components are connected and keep track of dependencies. Because of this, the change it makes are context aware, which is especially useful when working with large or complex projects. Next important capability is it can translate natural language into code. With cloud code, you don't need to think in terms of exact syntax all the time. You can simply explain what you want in normal English language. Whether it is adding a feature, fixing a bug or refactoring the code, cloud code understand your intent, plans the steps needed and then execute them making the whole development process feel much more intuitive. Cloud code is also designed to be flexible and extensible. There comes the fourth feature which is integration with git and developer tools. It can connect with external tools, APIs and services and it can work across different environments. This means you can build custom workflows and integrations on top of cloud code instead of being limited to basic coding tasks. Another feature is git workflow integration. Cloud code fits naturally into your existing git workflow. It can help create commits with meaningful messages, assist in resolving merge conflicts and support branching and version control operations. This makes collaboration smoother and helps keeps your repository clean and organized. And the sixth capability of cloud code is it works best as supervised agentic assistant. You can think of it like a junior developer. It executes stars based on your instruction but you stay in control. You review the changes, approve them and guide the direction which keeps the workflow both powerful and safe. We have covered the core capabilities of cloud code and it keeps getting better with ongoing updates from anthropic. Recently, cloud code received an important update that improves how it handles external tools and integration with MCP tool search. Previously, cloud code would preload multiple tools. It would load information about all of them up front, even if only one tool was actually needed. That added unnecessary context and slowed things down a bit. With the latest update, this behavior has changed. Cloud code now loads tool definition only when they are actually required, making the system more efficient and focused. The second update is faster and lighter execution. Since cloud code is no longer pre-loading all tool information, the system becomes more responsive. There is less overhead, fewer distraction in context, and quicker decision- making. In real world terms, this means better performance, especially when you are working with multiple tools on complex workflows. And finally, smarter context management. By not wasting prompt space on unused tools, clawed code preserves more room for what actually matters, your coding logic, reasoning and task specific. This leads to clearer reasoning, better outputs, and fewer chances of confusion or irrelevant responses. That brings us to the end of this video on the core capability of cloud code. I'll see you in the next video. Now, now you know what is cloud code and its features. Now let's understand how to access it. You can simply copy paste the installation command given for your operating system and run it directly in your terminal. You will find these commands neatly listed in the resource section attached with this video. If you prefer, you can just search for cloud code on the web. Click the first link from Entropic. And there you will see the installation code for your specific operating system, Windows, Mac or Linux. Now you can copy this code, open your terminal and paste the command here. Now you can see it will start installing. Now cloud code is successfully installed. Now to run cloud code you need to type cloud and here it will ask you for the theme. So I'll choose the dark mode. And now cloud code will ask you to login into your account. Either you can have the prom max team enterprise subscription or you can use your API for usage billing. So I'll click on the cloud account with subscription and I'll click on it. So now it will open in my browser. I'll authorize it. accept the cookies. Okay, now I can close this window. And here you can see I have successfully logged into my account. Now I press enter here and again I'll press enter and I'll go with the recommended settings and yes I'll proceed. So now you can see I can give my prompt and cloud code will do it for me. So now in the next part of the video we will move to the hands-on section and see how to actually work with cloud code in real projects. In this, we are going to build something fun, a mini Super Mario style 2D platformer game. This project will help you see how clot code can be used to translate a clear idea into an actual working application. Let's start by understanding the problem statement. Our goal in this project is to design and build a simple 2D Super Mario style platformer. Instead of jumping straight into the code, we will first look at a well- definfined prompt that clearly describes what the game should include. That prompt will guide everything we built next. This is the exact prompt we will be using for the project. It clearly outlines the game environment, a tile based layout, Mario standing on the ground block, a background sky with clouds, a question mark block, and a green pipe. It also specifies the core mechanics like left and right movement, jumping using arrow keys, gravity and collision with platforms. This kind of detailed prompt helps cloud code understand both visual elements and the game logic we want to implement. If you look closely, this prompt is doing a lot of work for us. It defines the game world, the visuals, the control, and even the physics behavior like gravity and collision. This is a great example of how well-written prompt can act like a mini design document for your project. Now that we are clear on the problem and the prompt, it's time to go hands-on. In the next step, we will use cloud code to start building its game, setting up the layout, adding movement and jumping mechanics, and bringing the Super Mario style experience to life step by step. So, let's get started. So, I've already created a folder named CL code. And here in this folder I'll open the terminal. I will initialize cloud code by using the command claude. And here I will copy paste the exact prompt I have showed you previously. So I'll just copy paste it and hit enter. Uh and using control T you can see the to-do list and the to-do list says create HTML structure with canvas element. Implement pixel at sprite rendering system. Build tile base level with ground clouds question block and pipe. add player character with keyboard controls. Implement gravity and collision detection and add game loop and polish. And now, so let's wait and see what cloud code gives us. So I'll just scroll it down and wait for cloud code to complete the implementation. So our Super Mario implementation is ready. Let's look into it. So here's the game and let's try it out. So I'll just hit jump here. go. Oh, and it's working properly. So, I'm really impressed how Cloud Code implemented this game so quickly. In the next part, we'll create something more than just a game. In this, we are going to build something very practical and impactful, a mock interviewer agent. This project focuses on fixing a real problem with how interview preparation works today. Let's start with the problem. Most interview preparation today doesn't reflect how real interview actually work. Candidates rely on generic questions, practice without meaningful feedback, and have no way to track whether they are improving. On top of that, real interviewers challenge your answer, ask follow-up questions, and probe weak areas. But traditional practice methods simply don't do that. Now, imagine a better approach. What if you had access to a senior interviewer available anytime? Someone who asks questions tailored to your background challenges weak answers with realistic follow-ups, scores your response, gives clear feedback, and track your improvement over time. That's the opportunity we are addressing with this project. In this project, we will build an AI powered mock interviewer agent. This agent will analyze your resume and the specific role you are targeting, stimulate realistic interview behavior, and evaluate your responses. The goal is not just practice but measurable improvement with every interview session. Now that the problem and solution are clear, it's time to go to hands-on. In the next step, we will use clot code to start building this mock interviewer agent step by step from understanding rums to generating questions and evaluating answers. So let's get started. So I have created an AI mock interviewer folder and now I'll initialize cloud in it using command claude. Now I have already crafted my prompt. You can get this prompt from the resource section of the course. So I'll just copy paste it now. And now I'll copy paste the prompt here in cloud code and hit enter. And here's my prompt. And let's see what cloud code will give us. Right now cloud code is thinking on the problem statement. Once the thinking process is done, it will create a checklist which we can see using the command D shortcut. So now our back end and front end are ready. So now let's try it out. It is asking for my name and my email address. Then I'll hit continue. So now let's paste my resume and the job description I'm targeting. Here's my resume. And uh and here's the job description. And I'll click on analyze and continue. Okay. So the current role is senior software engineer. Target role is senior software engineer. Experience is 7 years skills cloud and deps databases. So yeah, it's mapping everything correctly. Uh so let's click on start mock interview. So I'll choose mixed interview type and number of questions. Let number of questions be five. Let's click on start interview. Now it's asking me a question like tell me about a time when you had to make a difficult technical decision with incomplete information. How did you approach it? So I'll give some vague answer right now. I used Chad GBD. Click on submit. Interview feedback was I did like to hear more details. Can you elaborate on your specific role and the action you took? So yes, it is giving me a good feedback because I just said I use chart GPD. Okay. Then now I'll move to the next question. What was your specific contribution versus what the team did? My contribution was to add AI capabilities to the system and rest of the team was working on the front end and back end integration. So I'll submit this response and again I didn't elaborate much on the role and action I took. So it gave me the same feedback. So I'll just end the interview here and let's see the progress. So that's it. So yeah this AI mock interview is working well for me and I'm quite surprised how well Claude code did all this. In this we will clearly define the problem statement we are trying to solve and understand how clot code can help us move from an idea to a real production ready project. Let's start with the problem statement. In this project, we are building a productionready 3D portfolio website designed to convert recruiters interest into real job opportunities. Today, recruiters typically spend less than 7 seconds scanning candidates profile. That makes it difficult for static rums or traditional portfolios to stand out. A challenge is to create a single page mobile first experience that transform a LinkedIn profile and a resume into an immersive 3D portfolio. This portfolio should clearly showcase Genai skills, projects and real world impact while remaining easy to scan, visually engaging and recruiter friendly. The goal is not to replace a resume but to strengthen both the resume and LinkedIn presence. Now let's talk about how we craft a prompt for cloud code using LLM. Everything start with a clear idea what you want to build and why. Once you have that, you can use LLMs like Claude or Chart GPD to help craft a strong detailed prompt. The key step here is the context. By sharing your LinkedIn profile and resume, you allow model to understand your background skills and goals. This results in a prompt that is personalized to you instead of something generic. Now we will use this approach to start building the portfolio step by step using clot code. So let's get started. So I will open the folder I created clot code and then I'll open the terminal. Then I will initialize the clot code by using the command cloud. So now I will copy paste my prompt in clot code. and then hit enter. Here is my prompt. And now let's see the checklist. Here's the checklist. Uh it will create a HTML structure with meta tags, SEO and PWA manifest. Build a hero section with 3D neural network animation. About section with timeline and skill wheels. Build skills section with 3D constellation etc. Now let's wait and see what it creates. And now our portfolio is ready. Let's look into it. And so here's my name, Sam Jen. Neural network animation in the background. Uh, generative AI specialist building intelligent systems. Okay, here's the about section with core expertise and my experience. Here's the skills and expertise. Now, this looks nice. And here is my featured projects. Okay, nice. And here's the content and the tutorials I made in the past. Okay. And here's the let's connect section. So, I'm really impressed by how Claude Code created this portfolio. I really liked it. And that brings us to the end of this video. Thanks for watching and I'll see you in the next video. Hello and welcome back. In this video, we will clearly define the problem we are trying to solve and understand how clot code can help us with personalized exam preparation. In this course, we will explore clot code which powers AI to help personalize exam preparation. Making your notes exam ready with just few steps. So let's start with problem we are solving. Many students spends hours reading their notes but still struggle to identify what's actually important for their exams. They don't have access to personalized practice questions that match the exam patterns like J need or board exams and the generic study material they use don't align with those exam formats. This leads to inefficient preparation and a lot of exam anxiety. So here's how we are solving it. With the AI study assistant, you can easily upload your notes in PDF or text formats. Select the exam level like J need board gate etc. And then AI analyze the content and generate exam style questions, theory, numericals, coding, MCQs with answers and explanation and concise revision summaries to help you revise quickly. The AI study assistant offers several features. Smart analysis, auto detect subjects, topics and weight from your notes. It supports 12 exam levels like J, need, board, gate, etc. and difficulty grading. Classifying questions as easy, medium or hard. And the anti-h hallucination features ensure questions are strictly based on your notes. And finally, revision summaries provide quick highlights of key points for efficient revision. The EI study assistant is powered by NodeJS, TypeScript, and Express for back end. SQLite manages the database while cloud API from Enthropic powers the AI. The mobile app is built with Android Cotlin and PDF powers extract text from PDF nodes for analysis and question generation. Here's the architecture that ties it all together. The Android app communicates with REST API which processes the data through cloud AI and then data is stored in the SQL light database and used to generate personalized content for each user. This smooth integration between mobile AI and the database makes the entire system efficient and userfriendly. And here's the flow you will follow when using the system. First, upload your PDF notes. Then, select your exam level like J or NE. View content analysis based on your notes. Generate questions and MCQs based on analysis. And then finally, get a revision summary that highlights key point to focus on. Now that the problem is clear, it's time to go hands-on. So, I have created a folder named my app and here I will initialize cloud code using the command cloud. Now, cloud code is running. I have already crafted a prompt for my project. So I'll copy paste it here and let cloud code do the work. So here is my prompt. You are a senior engineer. Build a product ready AI study assistant agent and the specification the text tag behavior requirement and the clean modular architecture. So now I will allow cloud code to do the work. So I'll press yes and it will start making the changes now. Now let's see the to-do list uh using the command control T. And here it is. It will initialize project structure and dependencies, setup SQLite database schema, create exam level configuration model, etc. and etc. Now let's wait for cloud code to do the work. So our code is ready. So it created four folders for me. Android, Android app, backend and a database. Now cloud code has given me my code files in four different structured folders. Android, Android app, backend and database. Now we will upload this Android folder in Android Studio. So let's get into it. Now we have to start the gradal sync. Once the sync is completed, our app will be ready to be deployed. Now the gradal sync is done. Now to test our mobile app, we have two options. Either you can connect your mobile phone using USB and use the app or you can use a virtual device. For that you need to go to tools option and click on device manager. And here you will see an option to create a virtual device. I have already created mine and that is Pixel 9 Pro 4 fold. And now I'll show you the demo. So I'll click on running devices. Here I'll click on study assistant. I'll upload my sample documents here. uh and then click on them and here I'll choose undergraduate college theory numerical coding MCQ questions so I'll click on them and then I can find my questions MCQs and now let's try it out and now let's try out some questions so I'll click on this one the type is theory difficulty is easy and the topic is kyomatics So the question is define chynatics and explain why it is called geometry of motion. And the answer is chyomatic is a sub field of physics that describes the motion of points, bodies and systems etc. So yeah it's a good question. We got detailed answer with topic difficulty and type and question type. Now let's look at some hard questions. So here it is and the difficulty is hard and the topic is equation of motion. Oh yeah, that's a difficult one. Derive the second equation of motion S is equals to UT +/ A square using the graphical method. Okay. And we got a good example which derived the equation and this test understanding of equation of motion. Okay. We got the explanation too. That's nice. Now let's try some MCQs. And now by using cloud code, you can add even more features to this app. You can find my sister prompt in the source section of the course. And to make the app more productive, you can release an iOS app to expand access and make tools available to broader audience. Additionally, you can introduce performance analytics which will track your progress, highlight areas that needs improvement to optimize your preparation and further boost the learning. You can also incorporate space repetition, a proven technique to help you retain information more effectively. You can also consider the addition of collaborative study groups and voice-based QA features to further enhance your learning experience making it more interactive and engaging. In conclusion, this AI powered study assistance transform passive reading into active learning. It saves hours of manual questions and the exam specific preparation ensures better results with cloud code. Preparing for exams becomes easier and more efficient. In this we will clearly define the problem we are trying to solve and understand how clot code can help us with personalized exam preparation. In this course we will explore clot code which powers AI to help personalize exam preparation making your notes exam ready with just few steps. So let's start with problem we are solving. Many students spends hours reading their notes but still struggle to identify what's actually important for their exams. They don't have access to personalized practice questions that match the exam patterns like J need or board exams and the generic study material they use don't align with those exam formats. This leads to inefficient preparation and a lot of exam anxiety. So here's how we are solving it. With the AI study assistant, you can easily upload your notes in PDF or text formats. Select the exam level like J need board gate etc. and then AI analyze the content and generate exam style questions, theory, numericals, coding, MCQs with answers and explanation and concise revision summaries to help you revise quickly. The AI study assistant offers several features, smart analysis, auto detect subjects, topics and weight from your notes. It supports 12 exam levels like J, need, board, gate, etc. and difficulty grading, classifying questions as easy, medium or hard. And the anti-h hallucination features ensure questions are strictly based on your notes. And finally, revision summaries provide quick highlights of key points for efficient revision. The AI study assistant is powered by NodeJS, TypeScript, and Express for back end. SQLite manages the database while cloud API from Enthropic powers the AI. The mobile app is built with Android Cotlin and PDF powers extract text from PDF nodes for analysis and question generation. Here's the architecture that ties it all together. The Android art communicates with REST API which processes the data through cloud AI and then data is stored in the SQL light database and used to generate personalized content for each user. This smooth integration between mobile AI and the database makes the entire system efficient and userfriendly. And here's the flow you will follow when using the system. First upload your PDF notes. Then select your exam level like J or need. View content analysis based on your notes. Generate questions and MCQs based on analysis. and then finally get a revision summary that highlights key point to focus on. Now that the problem is clear, it's time to go hands-on. So I have created a folder named my app and here I will initialize cloud code using the command cloud. Now cloud code is running. I have already crafted a prompt for my project. So I'll copy paste it here and let cloud code do the work. So here is my prompt. You are a senior engineer. build a product ready AI study assistant agent and the specification the text tag behavior requirement and the clean modular architecture. So now I will allow cloud code to do the work. So I'll press yes and it will start making the changes now. Now let's see the to-do list uh using the command control t and here it is. It will initialize project structure and dependencies, set up SQL like database schema, create exam level configuration model, etc. and etc. Now let's wait for cloud code to do the work. So our code is ready. So it created four folders for me. Android, Android app, backend and a database. Now cloud code has given me my code files in four different structured folders. Android, Android app, backend and database. Now we will upload this Android folder in Android Studio. So let's get into it. Now we have to start the gradal sync. Once the sync is completed, our app will be ready to be deployed. Now the gradal sync is done. Now to test our mobile app, we have two options. Either you can connect your mobile phone using USB and use the app or you can use a virtual device. For that you need to go to tools option and click on device manager. And here you will see an option to create a virtual device. I have already created mine and that is Pixel 9 Pro 4. And now I'll show you the demo. So I'll click on running devices. Here I'll click on study assistant. I'll upload my sample documents. Here click on them. And here I'll choose undergraduate college theory numerical coding MCQ questions. So I'll click on them. And then I can find my questions MCQs. And now let's try it out. And now let's try out some questions. So I'll click on this one. The type is theory. Difficulty is easy. And the topic is chyatics. So the question is define chynyatics and explain why it is called geometry of motion. And the answer is chyomatic is a sub field of physics that describes the motion of points, bodies and systems etc. So yeah it's a good question. We got detailed answer with topic difficulty and type and question type. Now let's look at some hard questions. So here it is and the difficulty is hard and the topic is equation of motion. Oh yeah that's a difficult one. Derive the second equation of motion S is equals to UT +/ A square using the graphical method. Okay. And we got a good example which derived the equation and this test understanding of equation of motion. Okay. We got the explanation too. That's nice. Now let's try some MCQs and now by using cloud code you can add even more features to this app. You can find my sister prompt in the source section of the course. And to make the app more productive, you can release an iOS app to expand access and make tools available to broader audience. Additionally, you can introduce performance analytics which will track your progress, highlight areas that needs improvement to optimize your preparation and further boost the learning. You can also incorporate space repetition, a proven technique to help you retain information more effectively. You can also consider the addition of collaborative study groups and voice-based QA features to further enhance your learning experience making it more interactive and engaging. In conclusion, this AI powered study assistance transform passive reading into active learning. It saves hours of manual questions and the exam specific preparation and shows better results with cloud code. Preparing for exams becomes easier and more efficient. Thank you for completing this course. You have learned how to harness cloud code to create powerful applications and AI agents. From building your first agent to developing full-scale projects like 3D portfolio website and AI study assistant, you now have the tools to integrate AI powered workflows into real world solution. Keep experimenting with clot code and it will help you elevate your coding journey. Thank you.