<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Anushka Mishra</title>
    <description>The latest articles on DEV Community by Anushka Mishra (@anushka_mishra_324a5632b9).</description>
    <link>https://dev.to/anushka_mishra_324a5632b9</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3890074%2F86e7c673-6eb1-4ca8-b766-4b50d084cf4f.jpeg</url>
      <title>DEV Community: Anushka Mishra</title>
      <link>https://dev.to/anushka_mishra_324a5632b9</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/anushka_mishra_324a5632b9"/>
    <language>en</language>
    <item>
      <title>Confessions of a Level 0.1 Beginner: Surviving the Google Cloud Maze</title>
      <dc:creator>Anushka Mishra</dc:creator>
      <pubDate>Fri, 24 Apr 2026 06:53:35 +0000</pubDate>
      <link>https://dev.to/anushka_mishra_324a5632b9/confessions-of-a-level-01-beginner-surviving-the-google-cloud-maze-41jl</link>
      <guid>https://dev.to/anushka_mishra_324a5632b9/confessions-of-a-level-01-beginner-surviving-the-google-cloud-maze-41jl</guid>
      <description>&lt;p&gt;&lt;a href="https://dev.tourl"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;I wanted to build my very first AI Agent on Google Cloud. I logged in with high energy and clicked "Create Agent." Suddenly, the "Billing &amp;amp; Automation" screen popped up. My heart skipped a beat! I actually started sweating because it looked so serious. I knew Google gives $300 free credits, but seeing those "Auto-pay" words still scared the life out of me! I gathered my courage and tried to pay via UPI. It failed. I tried again. Failed again. A third time? Failed! Turns out, my Net Banking was turned off the whole time.You can tell how much of a beginner I am—I was ready to build a "Smart AI," but I didn't even know my bank's "ON" switch. &lt;br&gt;
You might be thinking, If she’s a Level 0.1 beginner who got scared of a billing screen, why is she writing a tutorial? Actually I wanted to test if an AI Agent is literally Simple  or it’s  just Marketing Hype. I’m not here to show off my coding skills (because I have none!). Actually I’m here to tell you if a science student can survive the Google Cloud maze without losing her mind.&lt;br&gt;
So If I can go from "Net Banking Fail" to building a bot, YOU CAN TOO. Let’s see what happens next &lt;br&gt;
Hey! As a fellow science student, I totally get that beginner's curiosity mixed with a bit of "what is even happening?" when looking at AI. Since I am  just starting to explore this ai world, let’s break down what this Vertex AI Agent actually is.&lt;br&gt;
Think of me as your college senior explaining this over a cup of tea.&lt;br&gt;
&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fhqg5eqsgbmpcol5de5r0.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fhqg5eqsgbmpcol5de5r0.png" alt="Here is a visual explanation of the ai agent" width="800" height="462"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;What exactly is this "AI Agent"?&lt;br&gt;
Let’s skip the boring definitions and get straight to it without any complex coding languages.&lt;br&gt;
Imagine AI is a super-smart encyclopedia (that’s basic AI). But an AI Agent? That’s an encyclopedia that has hands, a brain, and a "can-do" attitude.&lt;br&gt;
Here is the breakdown in my way :&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fkqlgnb1clws2659kznyd.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fkqlgnb1clws2659kznyd.png" alt="Here is the difference between normal ai  vs ai agent " width="800" height="437"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;It’s a "Do-er," not just a "Talker": Normal AI just answers questions. An AI Agent takes action. If I ask a normal AI about a trip, it gives me a list. An AI Agent says. I’ve checked the flights, looked at your calendar, and drafted the itinerary. It doesn't just chat; it works.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fsmsf7z03rksnrvx23gau.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fsmsf7z03rksnrvx23gau.png" alt="this image tells abut how normal ai is stand alone adviser and ai agent is a context aware partner " width="800" height="437"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;It’s like a "Goal-Oriented" friend : If I give it a destination, and it finds the map. Instead of me telling it every single step (like a baby), I give it a goal: "Make a quiz from this Biology chapter." The Agent figures out the PDF, pick the hard words, and create the questions by itself.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F7z2j2qqr5clxx5nppsi8.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F7z2j2qqr5clxx5nppsi8.png" alt="h this image is explaining how ai agents have its own brain logic" width="800" height="437"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;It has its own Reasoning: Agents don't just guess the next word. They think before they speak. They use a "Chain of Thought." It’s like the Agent is saying, "First, I need to open the book, then I need to find the Animal section, and finally, I will explain what a Tiger is." * The Ultimate "Lazy Student" Hack: For a beginner, an Agent is like having a personal assistant who never sleeps. It handles the "boring manual labor" (like searching 50 websites or organizing messy notes) so you can focus on the fun part—like being the "Boss" of your project.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fgcbph3clslwxc8lz2so5.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fgcbph3clslwxc8lz2so5.png" alt="image showing ai agents learning and adaptations " width="800" height="437"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;It Learns and Adapts: A basic chatbot forgets who I am  the moment I close the tab. But an Agent can be built to remember my  style. If I  like funny explanations, it learns to be funny. It’s not a one-size-fits-all tool; it becomes MY tool.&lt;br&gt;
How it improves our Day-to-Day Life&lt;br&gt;
You might think, "It’s just a bot," but here’s how it actually changes things:&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F4v5k31l2y7xu6jwwoogk.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F4v5k31l2y7xu6jwwoogk.png" alt="this image showing that how ai is a instant explorer, super fact checker and personalised learning body at the same time" width="800" height="437"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Imagine I am  at a forest suddenly a special kind of frog came in front of me and have a specific question like, "Can this specific frog's venom be used in medicine?" Instead of scrolling through 20 Google links, the Agent gives me  the exact answer instantly.&lt;br&gt;
If I  have grounded it in Google Search, it reduces "Hallucinations" it means when AI lies confidently. It’s like having a friend who always double-checks the textbook before speaking.&lt;br&gt;
As a student, I can ask it to explain complex zoology topics in "5th-grade English" or "Scientific Latin." It adapts to me.&lt;br&gt;
How it helps me as a Beginner&lt;br&gt;
 No-Code Entry&lt;br&gt;
I don't need to be a hardcore coder and also I am not. Vertex AI is "Low-Code/No-Code." Your "Logic" and "Creativity" are more important than knowing how to write Python. If anyone can explain a process in English, that person can build an Agent.&lt;br&gt;
 Logical Thinking&lt;br&gt;
Being a science student it helps  me here. Building an agent is just like a Science Experiment:&lt;br&gt;
    Hypothesis: "I want a bot that identifies animals."&lt;br&gt;
    Procedure: The many steps we will discuss.&lt;br&gt;
    Observation: Testing the bot.&lt;br&gt;
    Conclusion: Refining the bot's answers.&lt;br&gt;
Let’s see how I have made a chatbot &lt;br&gt;
After logging in and finished billing setup for the next 30 minutes, I was just a girl on a clicking mission. Click one thing, get confused, click another. I opened tabs I didn't understand and closed menus I couldn't name. It felt like I was mixing random chemicals in a lab, hoping I wouldn't accidentally start a fire!&lt;br&gt;
Finally, after a lot of "thinking and overthinking," things started to make sense. I finally found my way to the Agent Garden and the Agent Engine.&lt;br&gt;
 I stopped clicking randomly and started my first real task: The Prompt.&lt;br&gt;
So let’s jump in tutorial how I have  created my first own chatbot :&lt;br&gt;
First, I’m heading to the Google Cloud Console to enable the Vertex AI APIs.&lt;br&gt;
 I’m staring at the billing setup right now—it’s a total headache for students, but I’m pushing through it!&lt;br&gt;
 I’m clicking those three little lines (the menu) on the top left. I’m hunting for "Agent Builder." There are a many menus here, but I’m staying focused so I don’t get lost.&lt;/p&gt;

&lt;p&gt;I’m jumping straight into the "Agent Garden." Why start from zero? I’m picking a template to see exactly how the "pros" structure their bots.&lt;/p&gt;

&lt;p&gt;I’m clicking "Create Agent" now. In the Overview, I’m telling my bot exactly who it is: "You are a friendly Animal Expert."  The techy language here is a bit scary, but I’m keeping my eyes on the animal theme.&lt;/p&gt;

&lt;p&gt;I’m using the "Create" button on the left to map out the conversation. It feels like I'm writing a script for a play where I'm the director.&lt;/p&gt;

&lt;p&gt;I’m opening the "Prompt Gallery." I’m picking an animal-related prompt so I don’t have to sit here and type out 50 different rules myself. Smart work over hard work!&lt;/p&gt;

&lt;p&gt;I’m flipping the switch for "Grounding with Google Search." This is how I make sure my bot gives real-world facts instead of just hallucinating stories.&lt;/p&gt;

&lt;p&gt;I’m selecting "Structured Output." I want the info to look like a neat fact sheet, not a giant wall of text that nobody wants to read.&lt;/p&gt;

&lt;p&gt;I’m adjusting the "Token Limit." I’m teaching my bot to give quick, snappy answers.&lt;br&gt;
 I’m being careful here—if I set it too low, my bot might just stop talking mid-sentence!&lt;br&gt;
I’m hitting Save and clicking "Display." My "Paper Plan" is officially alive on the screen. Testing it right now... and oh my god, it actually works!&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fdkul3qekgq6p0ywaxmqg.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fdkul3qekgq6p0ywaxmqg.png" alt="Image summary of the steps mentioned above " width="800" height="436"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;I was determined to build a "Textbook Chatbot" to help me study.  After 30 minutes of clicking, exploring the "Garden," and trying to stay calm... I realized I needed a win. Sometimes, you have to crawl before you can run. Building a full textbook bot was like jumping into a deep ocean.I tried and I tried, but I ended up pivoting. Instead of biology notes, I built an Animal Information Chatbot. It wasn’t the "Textbook Genius" I imagined, but it was mine. It worked. I finally went from Level 0 to Level 1. &lt;br&gt;
Is AI for Beginners?&lt;br&gt;
I think yes it is. As  I started at Level 0.1 with sweaty palms and a "Net Banking Fail." I got lost in the "Agent Garden" and changed my project halfway through. Tech platforms might look like scary spaceships, but they are just tools. You don't need to be an expert to start; you just need to be brave enough to click. My chatbot might not be a textbook genius yet, but it’s a living proof that a Science student can survive the Google Cloud maze.&lt;br&gt;
• My Message: To everyone staring at the billing screen or feeling "lost" in the navigation bar—Don't quit. Your "Level 0.1" is just the beginning of something great.&lt;br&gt;
If I can do it, you definitely can. See you in the Cloud! &lt;/p&gt;

&lt;p&gt;This is a submission for the &lt;a href="https://dev.to/challenges/google-cloud-next-2026-04-22"&gt;Google Cloud NEXT Writing Challenge&lt;/a&gt;*&lt;/p&gt;

</description>
      <category>devchallenge</category>
      <category>cloudnextchallenge</category>
      <category>googlecloud</category>
      <category>ai</category>
    </item>
  </channel>
</rss>
