<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Bharadwajk on Medium]]></title>
        <description><![CDATA[Stories by Bharadwajk on Medium]]></description>
        <link>https://medium.com/@bharadwajk32?source=rss-deaff1d8b7cf------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*MIltEwcDWTVbllfz</url>
            <title>Stories by Bharadwajk on Medium</title>
            <link>https://medium.com/@bharadwajk32?source=rss-deaff1d8b7cf------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 17:42:20 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@bharadwajk32/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[☁️ Terraform Week: From .tf Files to Thinking Like an Infrastructure Engineer]]></title>
            <link>https://medium.com/@bharadwajk32/%EF%B8%8F-terraform-week-from-tf-files-to-thinking-like-an-infrastructure-engineer-de948d6faf76?source=rss-deaff1d8b7cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/de948d6faf76</guid>
            <dc:creator><![CDATA[Bharadwajk]]></dc:creator>
            <pubDate>Wed, 09 Sep 2026 13:26:22 GMT</pubDate>
            <atom:updated>2026-09-09T13:29:03.280Z</atom:updated>
            <content:encoded><![CDATA[<p>P.S. This post is part of the DevOps Micro Internship with Agentic AI Cohort 3 by <a href="https://www.linkedin.com/in/pravin-mishra-aws-trainer/">Pravin Mishra</a> (<a href="https://dmi.pravinmishra.com/">https://dmi.pravinmishra.com</a>) . You can begin your DevOps journey by joining the DMI waiting list. ( <a href="https://www.linkedin.com/safety/go/?url=https%3A%2F%2Flnkd.in%2Fd735Nsd7&amp;urlhash=awdv&amp;mt=qcQV2qn317G-nf9sbpffsLfSN2ZWQgvnC1UOxENJnpY9Ft3pSF1fAtAAq-ITH1oKZj6p6EAe-b-150SlRQXVaEQ8__kpd-8-ZFVIsIl1fVioVziolhjRdaWs4w&amp;isSdui=true">https://lnkd.in/d735Nsd7</a>)</p><p>View my progress here: <a href="https://dmi.pravinmishra.com/s/kachiraju.html">https://dmi.pravinmishra.com/s/kachiraju.html</a></p><p>This week of my <strong>DevOps Micro Internship</strong> was all about <strong>Terraform, AWS, and Agentic AI</strong>. 🛠️🤖</p><p>Until now, I had mostly worked with AWS by creating and configuring resources directly. Terraform changed that perspective.</p><p>Instead of asking:</p><blockquote><em>“Which button should I click in AWS?”</em></blockquote><p>I started asking:</p><blockquote><em>“How should this infrastructure be defined, reviewed, secured, and maintained as code?”</em></blockquote><p>And that turned out to be a much bigger shift than I expected. 😄</p><h3>🌱 Infrastructure as Code</h3><p>I started with the fundamental Terraform workflow:</p><p><strong>Write → Format → Validate → Plan → Review → Apply</strong></p><p>Commands like terraform init, fmt, validate, plan, and apply started making much more sense when I used them against real infrastructure.</p><p>One important lesson:</p><p><strong>terraform plan deserves as much attention as </strong><strong>terraform apply.</strong></p><p>apply makes the change.<br>plan tells you whether making that change is a good idea. 🔍</p><h3>🏗️ Building a Three-Tier AWS Architecture</h3><p>The major assignment was deploying my <strong>Book Review App infrastructure using Terraform</strong>.</p><p>The architecture followed:</p><p><strong>Internet → Public ALB → Web Tier → Internal ALB → Application Tier → RDS MySQL</strong></p><p>I worked with:</p><ul><li>🌐 VPC and public/private subnets across multiple AZs</li><li>⚖️ Public and internal Application Load Balancers</li><li>🖥️ Auto Scaling Groups</li><li>🔐 Security Groups with tier-to-tier access</li><li>🗄️ RDS MySQL with Multi-AZ and a read replica</li><li>🔑 Secrets Manager and IAM</li><li>📦 S3 for deployment artifacts</li></ul><p>One concept became especially clear:</p><p><strong>Private doesn’t simply mean “no one knows the IP.”</strong> 😄</p><p>The network itself should enforce who can communicate with whom.</p><p>The backend on port 3001 remained private, while MySQL on 3306 was accessible only from the application tier.</p><p>That changed how I look at Security Groups — from <strong>“which ports should I open?”</strong> to <strong>“who actually needs to talk to whom?”</strong></p><h3>🤖 Adding Agentic AI</h3><p>Another interesting part was using <strong>Agentic AI alongside Terraform</strong>.</p><p>AI assisted with:</p><ul><li>🧠 Infrastructure planning</li><li>🧩 Terraform implementation</li><li>🔐 Security reviews</li><li>🔍 Plan analysis</li><li>🛠️ Troubleshooting</li></ul><p>But there was an important rule:</p><p><strong>AI could analyze and recommend, but infrastructure changes still required human review.</strong></p><p>AI became the co-pilot — not the person with unlimited access to the cockpit controls. ✈️😄</p><h3>🔍 Learning Terraform Drift</h3><p>Then came another interesting problem:</p><p><strong>What happens when somebody manually changes a resource in AWS?</strong></p><p>The deployed infrastructure no longer matches what Terraform expects.</p><p>Hello, <strong>Terraform Drift!</strong> 👋</p><p>I intentionally introduced a harmless manual tag change and built a Bash drift-check workflow using:</p><p>terraform plan -detailed-exitcode</p><p>I also analyzed Terraform’s JSON plan with jq and checked for potentially dangerous changes such as:</p><ul><li>🗑️ Resource deletion</li><li>🔄 Resource replacement</li><li>🌍 Broad 0.0.0.0/0 exposure</li><li>⚠️ Unexpected modifications</li></ul><p>A reusable <strong>/tf-drift-review Agentic AI skill</strong> then helped turn the evidence into a plain-language risk assessment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EqYyFKUROuTSW_x-bVaNcA.png" /></figure><h3>🛡️ Putting a Guardrail Before terraform apply</h3><p>I also implemented a <strong>PreToolUse safety hook</strong> that blocks terraform apply when the latest drift review reports FAIL.</p><p>The workflow became:</p><p><strong>Detect → Collect Evidence → Analyze → Human Review → Act → Verify</strong></p><p>Interestingly, during testing the hook blocked an apply because the script reported a security issue.</p><p>Further investigation showed it was actually a <strong>false positive</strong> — the Terraform plan was only removing my test tag.</p><p>And that taught me another valuable lesson:</p><p><strong>Automation can be wrong too.</strong></p><p>A FAIL shouldn&#39;t always mean panic. 🚨</p><p>It should mean:</p><p><strong>“Stop and investigate before changing anything.”</strong></p><h3>🩺 Diagnose Before You Treat</h3><p>I started thinking about Terraform drift like an unexpected health-check result.</p><p>You don’t immediately start treatment because one number changed. First you understand <strong>what changed, why it changed, and how serious it is</strong>.</p><p>My workflow followed the same principle:</p><ul><li>🔍 <strong>Terraform + Bash</strong> → detect and collect evidence</li><li>📊 <strong>jq</strong> → extract important details</li><li>🤖 <strong>Agentic AI</strong> → analyze and explain risk</li><li>👨‍💻 <strong>Engineer</strong> → make the decision</li><li>🛑 <strong>Safety hook</strong> → prevent risky execution</li></ul><p>A little less <strong>“YOLO apply.”</strong> 😂<br>A lot more <strong>“show me the plan first.”</strong></p><h3>💡 My Biggest Takeaway</h3><p>At the beginning of the week, I thought Terraform was mainly about <strong>creating infrastructure using code</strong>.</p><p>Now I see the bigger picture:</p><p><strong>Code → State → Plan → Security → Drift → Review → Change → Verification</strong></p><p>Terraform isn’t just about automating infrastructure creation.</p><p>It’s about making infrastructure <strong>repeatable, reviewable, secure, and intentional</strong>.</p><p>And when Agentic AI joins the workflow, it can make engineers faster — but <strong>evidence, guardrails, and human judgment still matter.</strong> 🚀</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=de948d6faf76" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ DevOps Micro Internship — Week 07 Completed!]]></title>
            <link>https://medium.com/@bharadwajk32/devops-micro-internship-week-07-completed-ccf5af32409d?source=rss-deaff1d8b7cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/ccf5af32409d</guid>
            <dc:creator><![CDATA[Bharadwajk]]></dc:creator>
            <pubDate>Mon, 24 Aug 2026 13:43:25 GMT</pubDate>
            <atom:updated>2026-08-24T13:43:25.342Z</atom:updated>
            <content:encoded><![CDATA[<p>🚀 <strong>DevOps Micro Internship — Week 07 Completed!</strong></p><p>P.S. This post is part of the DevOps Micro Internship with Agentic AI Cohort 3 by <a href="https://www.linkedin.com/in/pravin-mishra-aws-trainer/">Pravin Mishra</a> (<a href="https://dmi.pravinmishra.com/">https://dmi.pravinmishra.com</a>) . You can begin your DevOps journey by joining the DMI waiting list. ( <a href="https://www.linkedin.com/safety/go/?url=https%3A%2F%2Flnkd.in%2Fd735Nsd7&amp;urlhash=awdv&amp;mt=qcQV2qn317G-nf9sbpffsLfSN2ZWQgvnC1UOxENJnpY9Ft3pSF1fAtAAq-ITH1oKZj6p6EAe-b-150SlRQXVaEQ8__kpd-8-ZFVIsIl1fVioVziolhjRdaWs4w&amp;isSdui=true">https://lnkd.in/d735Nsd7</a>)</p><p>View my progress here: <a href="https://dmi.pravinmishra.com/s/kachiraju.html">https://dmi.pravinmishra.com/s/kachiraju.html</a></p><p>This week felt very familiar… because the <strong>process stayed the same, but the cloud changed</strong>. 😄 After working through the AWS version, I repeated the same security and cloud engineering mindset on <strong>Azure</strong> — and quickly realized that changing aws to az does not magically remove troubleshooting!</p><p>This week I worked on real Azure tasks across <strong>compute, networking, storage, databases, security and Agentic AI</strong>:</p><ul><li>🌐 <strong>Deployed a static website using Azure Storage</strong></li><li>🏗️ <strong>Designed and deployed a three-tier Azure architecture</strong></li><li>📚 <strong>Deployed Book Review and Epic Book applications with MySQL in a three-tier architecture</strong></li><li>🔐 <strong>Configured private application and database tiers with controlled public access</strong></li><li>⚙️ <strong>Troubleshot Nginx, Application Gateway, NSG, networking, CORS, PM2 and API routing issues</strong></li><li>🤖 <strong>Built a Claude Code </strong><strong>/azure-audit skill to perform a read-only Azure security posture audit</strong></li><li>🔎 <strong>Found a real NSG security issue, fixed it manually, and re-ran the audit to verify the fix</strong></li></ul><p>The biggest learning for me was that these assignments are <strong>not about blindly following steps and ticking checkboxes</strong>. The real learning happens when something breaks, the error doesn’t make sense, you investigate it layer by layer, use AI where it actually helps, apply the fix yourself, and then prove that the fix worked.</p><p>AWS, Azure… different commands, same engineering mindset. 😄☁️</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VFL9ugrx-lQ1JFQa2gjxiw.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ccf5af32409d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From High Availability to AI Security Audits: My AWS Week 6 Journey ☁️]]></title>
            <link>https://medium.com/@bharadwajk32/from-high-availability-to-ai-security-audits-my-aws-week-6-journey-%EF%B8%8F-1302eb188036?source=rss-deaff1d8b7cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/1302eb188036</guid>
            <dc:creator><![CDATA[Bharadwajk]]></dc:creator>
            <pubDate>Sat, 15 Aug 2026 08:39:06 GMT</pubDate>
            <atom:updated>2026-08-15T08:39:06.584Z</atom:updated>
            <content:encoded><![CDATA[<p>P.S. This post is part of the DevOps Micro Internship with Agentic AI Cohort 3 by <a href="https://www.linkedin.com/in/pravin-mishra-aws-trainer/">Pravin Mishra</a> (<a href="https://dmi.pravinmishra.com/">https://dmi.pravinmishra.com</a>) . You can begin your DevOps journey by joining the DMI waiting list. ( <a href="https://www.linkedin.com/safety/go/?url=https%3A%2F%2Flnkd.in%2Fd735Nsd7&amp;urlhash=awdv&amp;mt=qcQV2qn317G-nf9sbpffsLfSN2ZWQgvnC1UOxENJnpY9Ft3pSF1fAtAAq-ITH1oKZj6p6EAe-b-150SlRQXVaEQ8__kpd-8-ZFVIsIl1fVioVziolhjRdaWs4w&amp;isSdui=true">https://lnkd.in/d735Nsd7</a>)</p><p>View my progress here: <a href="https://dmi.pravinmishra.com/s/kachiraju.html">https://dmi.pravinmishra.com/s/kachiraju.html</a></p><p>This week of my DevOps Micro Internship was all about connecting the dots — from <strong>building infrastructure</strong> to <strong>deploying an application</strong> and finally <strong>auditing its security</strong>.</p><h3>🏗️ Assignment 1: High Availability</h3><p>I built a highly available AWS architecture using <strong>EC2, Application Load Balancer, Auto Scaling Group, multiple Availability Zones, and RDS MySQL</strong>. More importantly, I tested failure scenarios to understand what actually happens when an instance goes down. This helped me move beyond simply knowing AWS services individually and understand how they work together.</p><h3>🔄 Assignment 2: Three-Tier Architecture</h3><p>Next, I deployed a <strong>Book Review application</strong> with separate <strong>frontend, backend, and database tiers</strong>. This was where troubleshooting became the real teacher! 😄 I dealt with <strong>ALB connectivity, HTTP vs HTTPS confusion, Nginx errors, PM2 process management, port conflicts, RDS connectivity, CORS issues, </strong><strong>localhost configuration, and the famous </strong><strong>/api/api duplicate path 🤦‍♂️</strong>.</p><p>The biggest lesson? An architecture diagram may look perfect, but getting every layer to communicate correctly is where the real engineering happens.</p><h3>🤖 Assignment 3: AI-Assisted AWS Security Audit</h3><p>Finally, I built a <strong>read-only AWS security and cost audit workflow</strong> using <strong>Bash, AWS CLI, and Claude</strong>. The Bash script gathered evidence, Claude analyzed the findings and suggested remediation, but <strong>I reviewed and executed the actual changes</strong>.</p><p>The audit identified issues such as <strong>SSH open to </strong><strong>0.0.0.0/0, S3 public-access settings, and unencrypted EBS volumes</strong>. I fixed the SSH exposure by restricting access to my own IP address and then verified the environment again.</p><h3>🎯 My Biggest Takeaway</h3><p>These three assignments formed one complete journey:</p><blockquote><strong><em>Build 🏗️ → Run 🔄 → Troubleshoot 🐛 → Audit 🔐 → Fix 🔧 → Verify ✅</em></strong></blockquote><p>I learned that DevOps is not just about knowing tools or AWS services. It’s about understanding <strong>how the complete system works together</strong>, what happens when it fails, and how to improve it safely.</p><p>And yes, I also learned that one extra /api can ruin your day. 😂</p><p><strong>Build it. Break it. Fix it. Audit it. Learn from it. 🚀</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1302eb188036" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Scrum Theory to an AI-Assisted Sprint: My Hands-On DevOps Journey]]></title>
            <link>https://medium.com/@bharadwajk32/from-scrum-theory-to-an-ai-assisted-sprint-my-hands-on-devops-journey-6d4b45cef993?source=rss-deaff1d8b7cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/6d4b45cef993</guid>
            <dc:creator><![CDATA[Bharadwajk]]></dc:creator>
            <pubDate>Sat, 08 Aug 2026 09:10:07 GMT</pubDate>
            <atom:updated>2026-08-08T09:10:07.673Z</atom:updated>
            <content:encoded><![CDATA[<p>P.S. This post is part of the DevOps Micro Internship with Agentic AI Cohort 3 by <a href="https://www.linkedin.com/in/pravin-mishra-aws-trainer/">Pravin Mishra</a> (<a href="https://dmi.pravinmishra.com/">https://dmi.pravinmishra.com</a>) . You can begin your DevOps journey by joining the DMI waiting list.(<a href="https://www.linkedin.com/safety/go/?url=https%3A%2F%2Flnkd.in%2Fd735Nsd7&amp;urlhash=awdv&amp;mt=qcQV2qn317G-nf9sbpffsLfSN2ZWQgvnC1UOxENJnpY9Ft3pSF1fAtAAq-ITH1oKZj6p6EAe-b-150SlRQXVaEQ8__kpd-8-ZFVIsIl1fVioVziolhjRdaWs4w&amp;isSdui=true">https://lnkd.in/d735Nsd7</a>)</p><p>View my progress here: <a href="https://dmi.pravinmishra.com/s/kachiraju.html">https://dmi.pravinmishra.com/s/kachiraju.html</a></p><p>When I started exploring DevOps as part of my Micro Internship, I initially looked at it as a collection of tools, practices, and technologies.</p><p>But one thing became clear very quickly:</p><p><strong>DevOps is not only about tools. It is also about how we plan, collaborate, deliver, measure, and improve.</strong></p><p>That is what led me to explore the <strong>DevOps Life Cycle, Waterfall, Agile, and Scrum</strong>.</p><p>Instead of stopping with theory, I decided to put Scrum into practice.</p><h3>From Theory to a Five-Day Sprint</h3><p>To understand how Scrum works in a real delivery scenario, I ran a <strong>five-day mini-Sprint using Jira</strong>.</p><p>The goal was to take a small website improvement from an idea all the way to a verified deployment.</p><p>I started with a <strong>Sprint Goal and User Story</strong>, then broke the work into smaller daily Sub-tasks.</p><p>Throughout the Sprint, I tracked:</p><ul><li>Daily Scrum updates</li><li>Jira comments and statuses</li><li>Git commits</li><li>Deployments to AWS EC2</li><li>Verification of the changes</li><li>Sprint progress through the Burndown Chart</li><li>Sprint retrospective</li></ul><p>This gave me an opportunity to experience the complete flow:</p><p><strong>Plan → Develop → Commit → Deploy → Verify → Improve</strong></p><p>It was interesting to see how something that sounds simple in theory becomes a continuous cycle when you actually perform it.</p><h3>A Second Challenge: Thinking Like a Scrum Team</h3><p>I then worked on another Scrum assignment using the <strong>Gotto Job website</strong>.</p><p>This time, the focus was on creating a value-ordered product backlog and planning a realistic Sprint.</p><p>I worked in <strong>Solo Mode</strong>, which meant I had to take on multiple responsibilities that would normally be distributed across a Scrum team.</p><p>I had to think like the:</p><p><strong>Product Owner → Scrum Master → Dev Lead → DevOps Lead</strong></p><p>I created and refined the backlog under an Epic, prioritized the work based on value, and estimated stories using <strong>Fibonacci Story Points — 1, 2, and 3</strong>.</p><p>From there, I planned Sprint 1 around a clear Sprint Goal and selected a realistic scope.</p><p>The deliverable itself was intentionally small: a <strong>UI-only improvement</strong> such as text, color, spacing, a label, or CTA.</p><p>The idea was not to build something huge.</p><p>It was to experience the <strong>delivery process</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ny6Ze5rGoxwbT2vMca3_Rw.png" /></figure><h3>Then I Introduced AI 🤖</h3><p>This is where the exercise became even more interesting for me.</p><p>After experiencing the Sprint manually, I introduced an <strong>AI skill</strong> into the process.</p><p>Instead of using AI simply to generate code, I used it as a kind of <strong>Sprint health-check assistant</strong>.</p><p>The AI reviews the Sprint and produces a <strong>triage report</strong>.</p><p>It looks at the available Sprint information and helps identify areas that may need attention.</p><p>This changes the role of AI from simply being a coding assistant to becoming something closer to a <strong>delivery feedback mechanism</strong>.</p><p>And this is where I connected it with another important concept:</p><h3>Gather → Verify → Act</h3><p>The AI-assisted process fits naturally into a simple continuous-improvement loop:</p><p><strong>Gather → Verify → Act → Repeat</strong></p><h3>1. Gather</h3><p>Collect the available Sprint information:</p><ul><li>Jira statuses</li><li>Story progress</li><li>Sub-task completion</li><li>Comments</li><li>Burndown information</li><li>Delivery evidence</li></ul><h3>2. Verify</h3><p>AI analyses the information and performs a <strong>Sprint health check</strong>.</p><p>It can help identify potential issues, gaps, inconsistencies, or areas that deserve attention.</p><p>The result is a <strong>triage report</strong> rather than just raw Sprint data.</p><h3>3. Act</h3><p>The team can then use those findings to decide what action is required.</p><p>For example:</p><ul><li>Is a story not progressing?</li><li>Is work getting carried over?</li><li>Is the Sprint scope realistic?</li><li>Is there a delivery or verification gap?</li><li>Is additional attention required?</li></ul><p>The important point is that AI is helping turn information into <strong>actionable feedback</strong>.</p><h3>4. Repeat</h3><p>After taking action, the process starts again.</p><p>New information is gathered, verified, acted upon, and reviewed.</p><p>That creates a continuous improvement loop rather than a one-time health check.</p><h3>What I Learned</h3><p>The biggest takeaway from these exercises was that <strong>Scrum and DevOps complement each other</strong>.</p><p>Scrum gives structure to how work is planned, prioritized, executed, and reviewed.</p><p>DevOps brings the engineering and delivery practices needed to turn that work into a working increment.</p><p>And AI can add another layer by helping us <strong>observe the delivery process, identify signals, and support better decisions</strong>.</p><p>The progression for me looked something like this:</p><p><strong>Learn the methodology → Practice the Sprint → Deliver the increment → Measure the Sprint → Introduce AI → Improve the feedback loop</strong></p><p>What started as a Scrum learning exercise became a practical experiment in combining <strong>Agile, DevOps, Jira, Git, AWS EC2, and AI-assisted delivery</strong>.</p><p>And perhaps the most important lesson was this:</p><blockquote><strong><em>The goal isn’t to automate everything. The goal is to create a better feedback loop so that we can learn and improve faster.</em></strong></blockquote><p>This experience has given me a much better understanding of why DevOps is often described as a culture of <strong>continuous improvement</strong>.</p><p>I’m looking forward to exploring where AI can take this workflow next. 🚀</p><p>#DevOps #Agile #Scrum #AI #Jira #AWS #EC2 #Git #DevOpsInternship #ContinuousImprovement #LearningByDoing</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6d4b45cef993" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ The Hidden Observer Wasn’t Alone: My Week Exploring Git, GitHub & AI]]></title>
            <link>https://medium.com/@bharadwajk32/the-hidden-observer-wasnt-alone-my-week-exploring-git-github-ai-2ca2de033782?source=rss-deaff1d8b7cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/2ca2de033782</guid>
            <dc:creator><![CDATA[Bharadwajk]]></dc:creator>
            <pubDate>Thu, 23 Jul 2026 04:27:23 GMT</pubDate>
            <atom:updated>2026-07-27T14:12:06.552Z</atom:updated>
            <content:encoded><![CDATA[<p>P.S. This post is part of the DevOps Micro Internship with Agentic AI Cohort 3 by <a href="https://www.linkedin.com/in/pravin-mishra-aws-trainer/">Pravin Mishra</a> (<a href="https://dmi.pravinmishra.com/">https://dmi.pravinmishra.com</a>) . You can begin your DevOps journey by joining the DMI waiting list. ( <a href="https://www.linkedin.com/safety/go/?url=https%3A%2F%2Flnkd.in%2Fd735Nsd7&amp;urlhash=awdv&amp;mt=qcQV2qn317G-nf9sbpffsLfSN2ZWQgvnC1UOxENJnpY9Ft3pSF1fAtAAq-ITH1oKZj6p6EAe-b-150SlRQXVaEQ8__kpd-8-ZFVIsIl1fVioVziolhjRdaWs4w&amp;isSdui=true">https://lnkd.in/d735Nsd7</a>)</p><p>View my progress here: <a href="https://dmi.pravinmishra.com/s/kachiraju.html">Bharadwaja Kachiraju — DMI Cohort 3 · Live</a></p><p>A few days ago, I met a silent observer.</p><p>It didn’t speak.</p><p>It didn’t modify my files.</p><p>It simply watched everything I did.</p><p>Its name was <strong>.git</strong>.</p><p>When I ran git init, nothing on my screen looked magical. A hidden folder quietly appeared inside my project. It wasn&#39;t part of my website, yet it immediately started remembering every change I made.</p><p>It reminded me of those mystery movies where someone in a black coat watches the hero from a distance. Most people don’t even notice the observer until they shine a flashlight on it.</p><p>For me, that flashlight was:</p><p>ls -a</p><p>Suddenly, the hidden observer revealed itself.</p><p>Curiosity took over.</p><p>I created a tiny project with just an index.html and a style.css file. Every time I changed something, I watched Git carefully.</p><p>Untracked.</p><p>Staged.</p><p>Committed.</p><p>Then another commit.</p><p>And another.</p><p>Using git status and git log, I wasn&#39;t just running commands anymore—I was watching Git build the history of my project one snapshot at a time.</p><p>At one point, I even joked to myself:</p><blockquote><strong><em>“.git probably doesn’t know it has another observer… me.”</em></strong><em> 😄</em></blockquote><p>Instead of writing code, I was busy observing the observer.</p><p>Once I understood how Git worked locally, another question appeared.</p><p><em>“What happens when hundreds of developers are working on the same project?”</em></p><p>That question led me to GitHub.</p><p>I learned that collaboration isn’t about everyone editing the same file at the same time.</p><p>Instead, every developer gets their own safe workspace.</p><p>I forked an open-source repository.</p><p>Created a feature branch.</p><p>Made my changes.</p><p>Pushed them to <strong>my</strong> repository.</p><p>Then came something that confused me for a moment.</p><p><strong>Pull Request.</strong></p><p>I kept thinking…</p><p><em>“I’m pushing my code. Why am I creating something called a Pull Request?”</em></p><p>Then the perspective changed.</p><p>I wasn’t asking GitHub to push my code anywhere.</p><p>I was requesting the maintainers:</p><blockquote><strong><em>“If you like my changes, please pull them into your project.”</em></strong></blockquote><p>Sometimes understanding a concept is simply about changing your point of view.</p><p>Just when I thought Git was only about tracking history and collaboration, another lesson appeared.</p><p>Security.</p><p>This assignment flipped my expectations completely.</p><p>For the first time…</p><p>I actually wanted my commit to fail.</p><p>Sounds strange, doesn’t it?</p><p>Normally a rejected commit is frustrating.</p><p>But this time I intentionally placed something that looked like a secret inside my code.</p><p>Git refused to commit it.</p><p>Instead of feeling disappointed…</p><p>I felt relieved.</p><p>The security checkpoint had done exactly what it was supposed to do.</p><p>It reminded me that the best security incident is the one that never reaches the repository.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*M-4eCHwpyH1VQ0bA2QlZFQ.png" /></figure><p>But this checkpoint wasn’t working alone.</p><p>Standing beside it was my AI co-pilot.</p><p>Both looked at the exact same changes.</p><p>Yet they behaved very differently.</p><p>The <strong>pre-commit hook</strong> was like a strict security guard.</p><p>It followed fixed rules.</p><p>If something matched a known dangerous pattern, it immediately stopped me.</p><p>No discussions.</p><p>No exceptions.</p><p>The AI reviewer behaved more like an experienced senior engineer.</p><p>It didn’t just look for rules.</p><p>It looked for context.</p><p>It noticed leftover debug statements.</p><p>It explained why something could become a problem.</p><p>It suggested improvements instead of simply saying “No.”</p><p>Of course…</p><p>I didn’t hand over complete control.</p><p>I gave it permission to inspect my work.</p><p>Not permission to change it.</p><p>After all…</p><p><strong>Trust is good. Least privilege is better.</strong> 😄</p><p>I wanted advice.</p><p>The final decision still belonged to the engineer.</p><p>Something else quietly connected all these assignments.</p><p>Whether I was watching Git…</p><p>Opening a Pull Request…</p><p>Blocking a secret…</p><p>Or asking AI to review my work…</p><p>The same pattern kept appearing.</p><p><strong>Gather → Analyze → Human Act → Verify</strong></p><p>Git gathered changes.</p><p>The pre-commit hook analyzed them.</p><p>The AI added context.</p><p>I made the final decision.</p><p>Then everything was verified before moving forward.</p><p>This week taught me much more than Git commands.</p><p>It taught me why Git remembers every change, why branches exist, why Pull Requests are named the way they are, why security checks stop mistakes before they spread, and why AI works best as a thoughtful reviewer — not an engineer with unlimited permissions.</p><p>The more I learn, the more I realize that every command in Git has a purpose, every safety check has a story, and every assignment is less about memorizing syntax and more about understanding the ideas behind it.</p><p>The hidden observer that I met at the beginning of the week turned out not to be alone after all. It was part of a much larger team — developers, security checks, AI reviewers, and finally, the engineer making the last call.</p><p>And I can’t wait to meet the next member of that team.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2ca2de033782" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How I Joined the DevOps Micro Internship]]></title>
            <link>https://medium.com/@bharadwajk32/how-i-joined-the-devops-micro-internship-409cace6fcd4?source=rss-deaff1d8b7cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/409cace6fcd4</guid>
            <dc:creator><![CDATA[Bharadwajk]]></dc:creator>
            <pubDate>Sat, 18 Jul 2026 05:22:10 GMT</pubDate>
            <atom:updated>2026-07-18T05:23:14.192Z</atom:updated>
            <content:encoded><![CDATA[<p>P.S. This post is part of the DevOps Micro Internship with Agentic AI Cohort 3 by <a href="https://www.linkedin.com/in/pravin-mishra-aws-trainer/">Pravin Mishra</a>. You can begin your DevOps journey by joining the DMI waiting list. ( <a href="https://www.linkedin.com/safety/go/?url=https%3A%2F%2Flnkd%2Ein%2Fd735Nsd7&amp;urlhash=awdv&amp;mt=q8TGp4Ai7csRSBfC3tjE1F9JGC_lbFgdMNRyAjI1BfMOWuHikwgE7OxJJ6Pma4cnZjtjm2_3zQvfHk1ZU0lnqge5IkVIcoi10-9cAEcmDYa1qnp1YwZp2jT9kw&amp;isSdui=true">https://lnkd.in/d735Nsd7</a>)</p><p>In my previous posts, I shared my learning journey in the <strong>DevOps Micro Internship</strong> by <strong>Pravin Mishra</strong> and the concepts I’ve learned so far. Several people have asked me how they can join the internship, so I thought I’d write a post explaining the process.</p><p>First, let me share how I joined the program.</p><p>The selection process I followed was straightforward:</p><ol><li>I filled out the registration form.</li><li>After registering, I was given a few assignments to complete and submit.</li><li>Once my assignments were reviewed, I was shortlisted for an interview.</li><li>The interview was not about testing deep technical knowledge. Instead, it focused on my mindset, willingness to learn, consistency, and commitment to completing the program.</li><li>After successfully clearing the interview, I was added to the internship community and officially became part of the cohort.</li></ol><p>From what I understand, the onboarding process for the next internship may be slightly different. So, if you’re interested in joining, I recommend keeping an eye on the official Discord channel for the latest announcements and instructions which I added in PS section at the top.</p><p>To help future applicants, I’m also attaching the assignment that I completed during my selection process. Hopefully, it gives you an idea of what to expect and helps you prepare.</p><p><a href="https://www.linkedin.com/posts/bharadwaja-kachiraju-78a45598_cohort-assignment-bharawaja-kachirajudocx-activity-7464980357029953537-oJEs?utm_source=share&amp;utm_medium=member_desktop&amp;rcm=ACoAABS2KxoBOPNTBIxog_qhN1vz4HLYmnjgQPY">https://www.linkedin.com/posts/bharadwaja-kachiraju-78a45598_cohort-assignment-bharawaja-kachirajudocx-activity-7464980357029953537-oJEs?utm_source=share&amp;utm_medium=member_desktop&amp;rcm=ACoAABS2KxoBOPNTBIxog_qhN1vz4HLYmnjgQPY</a></p><p>If you have any questions about my experience or the application process, feel free to reach out. I’ll be happy to help wherever I can.</p><p>I wish everyone who plans to apply the very best. Stay curious, stay consistent, and keep learning!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=409cace6fcd4" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ From Learning Linux Commands to Building an AI Operations Co-Pilot]]></title>
            <link>https://medium.com/@bharadwajk32/from-learning-linux-commands-to-building-an-ai-operations-co-pilot-e48676eff771?source=rss-deaff1d8b7cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/e48676eff771</guid>
            <dc:creator><![CDATA[Bharadwajk]]></dc:creator>
            <pubDate>Fri, 17 Jul 2026 13:30:58 GMT</pubDate>
            <atom:updated>2026-07-23T04:12:07.518Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>🚗 From Learning Linux Commands to Building an AI Operations Co-Pilot</strong></p><p>P.S. This post is part of the DevOps Micro Internship with Agentic AI Cohort 3 by <a href="https://www.linkedin.com/in/pravin-mishra-aws-trainer/">Pravin Mishra</a> (<a href="https://dmi.pravinmishra.com/">https://dmi.pravinmishra.com</a>) . You can begin your DevOps journey by joining the DMI waiting list. ( <a href="https://www.linkedin.com/safety/go/?url=https%3A%2F%2Flnkd%2Ein%2Fd735Nsd7&amp;urlhash=awdv&amp;mt=qcQV2qn317G-nf9sbpffsLfSN2ZWQgvnC1UOxENJnpY9Ft3pSF1fAtAAq-ITH1oKZj6p6EAe-b-150SlRQXVaEQ8__kpd-8-ZFVIsIl1fVioVziolhjRdaWs4w&amp;isSdui=true">https://lnkd.in/d735Nsd7</a>)</p><p><strong>The week I stopped learning commands and started thinking like an engineer.</strong></p><p>There are weeks where you learn a few commands.</p><p>There are weeks where you learn a new technology.</p><p>And then there are weeks where multiple concepts suddenly connect together and form a bigger picture.</p><p>This was one of those weeks.</p><p>A few weeks ago, Linux commands looked like random spells from the Harry Potter universe.</p><p>I was typing them with the same confidence as someone entering a WiFi password they copied from a sticky note.</p><p>Eventually something changed. Those commands stopped being commands. They became tools. And tools started solving problems.</p><p><strong>Chapter 1: The Bookstore That Didn’t Exist But Had Customers To Serve</strong></p><p>My learning philosophy lately has been simple:</p><p><em>Don’t learn technology in isolation.</em> <em>Put it to work.</em> Instead of reading Linux theory, I deployed a real website.</p><p>Well… a fictional business with very real operational problems.</p><p>Meet <strong>Epic Reads</strong>, an online bookstore.</p><p>Imagine this:</p><p>It’s midnight. A customer wants to buy <em>Atomic Habits</em> because tomorrow morning is definitely going to be the start of their new life. They open the website. The website is down. The customer loses motivation. The bookstore loses a sale. James Clear loses a reader.</p><p>Human civilization moves backwards slightly:).</p><p>My job was simple:</p><p>Keep Epic Reads alive.</p><p>Using an Ubuntu server running on AWS EC2 and Nginx, I deployed the website and started operating it like it was a production application.</p><p>That involved:</p><ul><li>Launching an EC2 instance</li><li>Connecting through SSH</li><li>Configuring Nginx</li><li>Managing Linux files and directories</li><li>Verifying deployment</li><li>Monitoring services</li><li>Validating HTTP responses</li></ul><p>Suddenly Linux wasn’t theory anymore.</p><p>It became infrastructure.</p><p><strong>What users see</strong></p><p><em>“A bookstore website.”</em></p><p><strong>What engineers see</strong></p><ul><li>Linux processes</li><li>File systems</li><li>Permissions</li><li>Network sockets</li><li>Web servers</li><li>Logs</li><li>Services</li><li>Ports</li></ul><p>Users see books.</p><p>Engineers see moving parts.</p><p><strong>Chapter 2: Good Cop, Bad Cop</strong></p><p>After deployment came operations. This was my favorite part.</p><p>I decided to play both roles.</p><p><strong>Bad Cop</strong></p><p>I intentionally broke things. Because production eventually does.</p><p>First incident:</p><p>I removed a semicolon from an Nginx configuration file. One missing character. Entire configuration broken. Thousands of engineers around the world just nodded while reading that sentence.</p><p>Second incident:</p><p>I emptied the Nginx web root directory. Nginx was running. Configuration was valid. Server was healthy. But the application was gone.</p><p>The result:</p><p>HTTP/1.1 500 Internal Server Error</p><p>The infrastructure equivalent of:</p><p><em>“The restaurant is open but there is no food.”</em></p><p><strong>Good Cop</strong></p><p>The good cop entered the scene.</p><p>First checks:</p><p>systemctl status nginx</p><p>Then:</p><p>nginx -t</p><p>Then:</p><p>journalctl -u nginx</p><p>Then:</p><p>tail -f /var/log/nginx/error.log</p><p>The syntax error was fixed. The website still wasn’t working.</p><p>More investigation.</p><p>Eventually the logs revealed the real issue: There was simply nothing left to serve.</p><p>Fortunately, the good cop believed in backups. Files restored.Service recovered.</p><p>Customers returned. Epic Reads survived another day</p><p><strong>Chapter 3: Learning To Drive Linux</strong></p><p>By this point I knew a lot of commands.But something felt wrong.I was repeating myself.</p><p>Again. And again. And again. Checking services. Checking disk. Checking memory. Checking ports. Checking logs. It felt like manually driving through city traffic every single day. You can do it. Nothing wrong with it. But eventually you start wishing for cruise control.</p><p><strong>Chapter 4: Bash Became My Cruise Control</strong></p><p>I started learning Bash scripting. Initially it looked harmless. echo “Hello World”</p><p>Classic mistake. Every developer knows that “Hello World” is the gateway drug. A few hours later you’re writing functions and debugging loops at 1 AM.</p><p>Slowly I introduced:</p><ul><li>Variables</li><li>Arrays</li><li>Loops</li><li>Conditions</li><li>Functions</li></ul><p>Each concept reduced repetition. Repeated values became variables. Repeated commands became loops. Repeated logic became functions.</p><p>Eventually:</p><p>./health-check.sh replaced pages of repetitive commands. The destination stayed the same. The driver remained responsible. But the repetitive work disappeared. That is what automation should do. Not replace engineers. Free engineers.</p><p><strong>Chapter 5: Giving Cruise Control an AI Co-Pilot</strong></p><p>Then came the fun part. I added another layer. Agentic AI. I created a reusable Claude Code skill called:</p><p>/linux-triage</p><p>The skill used a read-only Bash triage script to collect health information from Ubuntu and Nginx. Then came the experiment. I deliberately stopped Nginx. Chaos engineering on a budget.</p><p>The workflow looked like this:</p><p><strong>Step 1 — Gather</strong></p><p>Bash collected facts:</p><ul><li>Is Nginx running?</li><li>Is port 80 listening?</li><li>Is HTTP reachable?</li><li>What do the logs say?</li></ul><p><strong>Step 2 — Analyze</strong></p><p>Agentic AI analyzed the evidence.</p><p>Not assumptions.</p><p>Evidence.</p><p><strong>Step 3 — Human Act</strong></p><p>AI suggested:</p><p>sudo systemctl start nginx</p><p>But the human approved the action.</p><p>The engineer remained in control.</p><p><strong>Step 4 — Verify</strong></p><p>Run the checks again.</p><p>Confirm recovery.</p><p>Close the incident.</p><p>Simple.</p><p>Elegant.</p><p>Practical.</p><p><strong>The Observation That Changed How I Think About AI</strong></p><p>The biggest lesson from this experiment was surprisingly simple:</p><p><em>Bash collects facts.</em></p><p><em>Agentic AI analyzes facts.</em></p><p><em>Engineers approve actions.</em></p><p><em>Systems are verified again.</em></p><p>Or in simpler terms:</p><ul><li>Bash became the eyes.</li><li>AI became the analyst.</li><li>Engineers remained the decision makers.</li></ul><p>That feels less like replacing engineers.</p><p>And more like giving engineers a co-pilot.</p><p><strong>Personal Development Lessons</strong></p><p>This week wasn’t really about Linux.</p><p>Or Bash.</p><p>Or Nginx.</p><p>Or AI.</p><p>It was about changing the way I learn.</p><p>I stopped asking:</p><p><em>“How does this command work?”</em></p><p>and started asking:</p><p><em>“What problem does this solve?”</em></p><p>That single shift made learning faster and more enjoyable.</p><p>Because nobody wakes up wanting to learn journalctl.</p><p>People want to solve incidents.</p><p>journalctl just happens to be part of the story.</p><p><strong>Final Thoughts</strong></p><p>A few days ago I was learning Linux commands individually.</p><p>Today I have:</p><ul><li>Deployed applications</li><li>Operated web servers</li><li>Simulated incidents</li><li>Automated operational tasks</li><li>Added Agentic AI into the workflow</li></ul><p>The interesting part isn’t any individual technology.</p><p>It’s watching them connect.</p><p>Linux became automation.</p><p>Automation became operations.</p><p>Operations became intelligence.</p><p>And this journey feels like it’s only getting started.</p><p><strong><em>Automation reduces repetitive work.</em></strong></p><p><strong><em>Agentic AI reduces repetitive investigation.</em></strong></p><p><strong><em>Engineers still own the decisions.</em></strong></p><p>If you made it this far, thank you for reading.</p><p>Now if you’ll excuse me, I need to go deliberately break another server in the name of education.</p><p>It’s called a lab environment. The lawyers insisted I add that sentence.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*63g_q0lD7yMoxKaOCCGNpQ.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e48676eff771" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DevOps Micro Internship — Week 2 Completed: My Reflection]]></title>
            <link>https://medium.com/@bharadwajk32/devops-micro-internship-week-2-completed-my-reflection-0c74de9d3c18?source=rss-deaff1d8b7cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/0c74de9d3c18</guid>
            <dc:creator><![CDATA[Bharadwajk]]></dc:creator>
            <pubDate>Fri, 10 Jul 2026 14:25:40 GMT</pubDate>
            <atom:updated>2026-07-10T14:25:40.520Z</atom:updated>
            <content:encoded><![CDATA[<p>Week 2 of the DevOps Micro Internship was one of the most exciting learning experiences so far. This week introduced me to the world of <strong>Agentic AI</strong> and how it can reshape the way we approach DevOps and infrastructure management.</p><p>Until recently, my understanding of AI in engineering was mostly limited to writing prompts and receiving answers. This week completely changed that perspective.</p><p>I learned that the real power of AI lies not in asking better questions, but in designing better systems that allow AI agents to operate with the right context, capabilities, and boundaries.</p><p>One of the major concepts we explored was the <strong>Agentic AI loop</strong> — the process that enables an AI agent to observe, reason, act, and learn while working toward a goal.</p><p>We also explored several important concepts in Claude Code that make this possible in real-world projects:</p><ul><li><strong>Claude.md</strong> for providing project context, conventions, constraints, and instructions.</li><li><strong>Skills</strong> to get the right capabilities from the LLM for specific tasks.</li><li><strong>Subagents</strong> to separate responsibilities and keep context clean.</li><li><strong>MCP (Model Context Protocol)</strong> servers to connect agents with real-time data sources and external services.</li><li><strong>Hooks</strong> to introduce checkpoints, validations, and safeguards before actions are executed.</li><li><strong>Memory</strong> to allow agents to retain project decisions and context across sessions.</li></ul><p>The biggest realization for me was that AI is not about replacing engineers. Instead, it is about augmenting engineers by handling repetitive work while humans continue to provide direction and architecture.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cFsuVbaaEHI7muGu3F0BNA.png" /></figure><p>An AI agent can deploy infrastructure, generate code, analyze configurations, and automate workflows. However, if the context, instructions, or constraints are wrong, even a highly capable AI agent can move confidently in the wrong direction.</p><p>This led me to an interesting parallel with life itself.</p><p>Distractions in life behave very much like incorrect context in an AI system. They can slowly move us away from our intended destination without us realizing it. Just as an AI agent requires an architect to define the system boundaries and guide it toward the correct objective, we also need a clear architecture for our own lives to guide our decisions and actions.</p><p>The agent can execute.</p><p>The architect decides the destination.</p><p>That was my biggest takeaway from Week 2.</p><p>I am excited to continue exploring the intersection of <strong>AI, DevOps, and automation</strong>, and to better understand how these technologies can help engineers focus less on repetitive tasks and more on solving meaningful problems.</p><p>P.S. This post is a part of DevOps Micro Internship with Agentic AI Cohort-3 by Pravin Mishra. You can start your DevOps journey by joining this Discord community ( [<a href="https://discord.pravinmishra.com/%5d(https:/discord.pravinmishra.com/)">https://discord.pravinmishra.com/](https://discord.pravinmishra.com/)</a> ).</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0c74de9d3c18" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[BK: From a Telugu-medium student in rural Telangana to a technology leader who inspired others to…]]></title>
            <link>https://medium.com/@bharadwajk32/bk-from-a-telugu-medium-student-in-rural-telangana-to-a-technology-leader-who-inspired-others-to-8e7d2332eaaf?source=rss-deaff1d8b7cf------2</link>
            <guid isPermaLink="false">https://medium.com/p/8e7d2332eaaf</guid>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[internships]]></category>
            <category><![CDATA[learning-devops]]></category>
            <category><![CDATA[pravin-mishra]]></category>
            <dc:creator><![CDATA[Bharadwajk]]></dc:creator>
            <pubDate>Wed, 01 Jul 2026 10:01:50 GMT</pubDate>
            <atom:updated>2026-07-02T03:55:48.238Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>BK: From a Telugu-medium student in rural Telangana to a technology leader who inspired others to dream bigger</strong></p><p>P.S. This post is a part of DevOps Micro Internship with Agentic AI Cohort-3 by <a href="https://www.linkedin.com/in/pravin-mishra-aws-trainer/">Pravin Mishra</a>. You can start your DevOps journey by joining this <a href="https://discord.pravinmishra.com/">Discord community</a> ( <a href="https://discord.pravinmishra.com/">https://discord.pravinmishra.com/</a></p><p>Just a few years ago, Bharadwaja Kachraju aka BK set out to redefine what was possible for someone from his background. Raised in a rural part of Telangana, he completed his schooling in Telugu medium before transitioning to English medium for higher education. That transition was one of the biggest challenges of his life. He spent more time understanding the language than understanding the subjects and often questioned whether he would ever be confident enough to build a successful career or clear job interviews.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FH6NKCMF3QngK0lUPurmIw.jpeg" /></figure><p>His first breakthrough came when he secured an entry-level role in SAP after nearly two years of persistent effort. Instead of treating that as the destination, he used it as the foundation for continuous growth. Over the following years, he expanded his expertise beyond SAP by building strong skills in DevOps, cloud technologies, automation, and modern IT practices. He earned industry-recognized certifications, built a public GitHub portfolio showcasing automation projects and learning labs, and documented his learning journey through technical articles and blogs.</p><p>With more than a decade of experience in the IT industry, BK became known not only for solving complex technical problems but also for simplifying them for others. He led cross-functional initiatives, mentored junior engineers, and played a key role in modernizing infrastructure and delivery processes within his organization. His ability to combine technical depth with empathy earned him the trust of both colleagues and leadership.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/511/1*g4gy6Ol8yOXnEb0o4KES3g.png" /></figure><p>Outside his professional responsibilities, BK contributed actively to the technology community. He mentored students and early-career professionals, especially those from rural and non-English-medium backgrounds. Through talks, online content, and one-on-one mentoring, he demonstrated that consistent effort matters more than where a person begins.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1015/1*tyBbHHhPey9-XMXZWZbC1g.png" /></figure><p>Today, many professionals seek BK’s guidance not only for technical challenges but also for career development and personal growth. His journey has become living proof that language, background, and circumstances do not define a person’s potential. With discipline, continuous learning, and perseverance, he built the career he once doubted was possible and inspired countless others to believe that they could do the same.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8e7d2332eaaf" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>