<?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 TechDredd on Medium]]></title>
        <description><![CDATA[Stories by TechDredd on Medium]]></description>
        <link>https://medium.com/@TechDredd?source=rss-1f9658ed51d6------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*sFqIBraNURLK3L9DCjRCcA.png</url>
            <title>Stories by TechDredd on Medium</title>
            <link>https://medium.com/@TechDredd?source=rss-1f9658ed51d6------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 07:30:27 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@TechDredd/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[Jira Must Die: A 2026 Manifesto Against Software’s Favorite Torture Device]]></title>
            <link>https://medium.com/@TechDredd/jira-must-die-a-2026-manifesto-against-softwares-favorite-torture-device-7021e7e32fac?source=rss-1f9658ed51d6------2</link>
            <guid isPermaLink="false">https://medium.com/p/7021e7e32fac</guid>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[jira]]></category>
            <dc:creator><![CDATA[TechDredd]]></dc:creator>
            <pubDate>Thu, 01 Oct 2026 13:27:22 GMT</pubDate>
            <atom:updated>2026-10-01T13:27:22.294Z</atom:updated>
            <content:encoded><![CDATA[<p>It is the year 2026. We have <strong>autonomous rockets</strong> landing on ocean barges, <strong>fusion testbeds</strong> breaching net-energy parity, and AI models capable of <strong>scaffolding entire microservices</strong> in seconds.</p><p>And yet, every single morning, five hundred brilliant, highly paid engineers at my company gather around their screens, <strong>sigh in collective defeat</strong>, and wait several seconds for a single Jira page to render.</p><p>At the start of the year, I had the distinct displeasure of joining the Jira board at a new company. Five hundred top-tier engineers, a modern stack, an ambitious roadmap. And then I was handed my login to our Jira board.</p><p>I was told, with unironic enterprise enthusiasm, that the company had recently completed a massive migration from an on-premise Jira server to Atlassian Cloud. <em>“It’s so much faster now!”</em> they <strong>beamed</strong>. And to be fair, they were right: page load times had been successfully upgraded from “enough time to brew a fresh espresso” down to “enough time to deeply contemplate your career choices.”</p><p>I fucking hate Jira. I hate it with every fiber of my engineering soul. I hate that I feel like <strong>I’m working for a tool</strong> that’s supposed to make my life easier. I hate that every company still thinks it’s a vital organ of modern software development simply because it was the default option twenty years ago.</p><p>It’s time to call it what it is: an <strong>administrative parasite</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Y7j-cSmb56295nUC" /><figcaption>Photo by <a href="https://unsplash.com/@hellkisa?utm_source=medium&amp;utm_medium=referral">Alexander Zabrodskiy</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>UX Designed by Monkeys Fighting Over a Banana</h3><p>Let’s talk about the user interface. Who at Atlassian is actually allowed to design this thing? It feels like a committee of monkeys pulling a single banana in every direction imaginable, while a product director screams <em>“STREAMLINE IT!”</em> from the back of the room.</p><p>Every six months, Atlassian unveils a brand-new “redesign” that somehow manages to make the tool heavier, slower, and more opaque than the last iteration. They hide essential buttons behind three layers of nested sub-menus, collapse custom fields into an endless void of accordion tabs, and replace clear text actions with ambiguous micro-icons that look like <strong>hieroglyphics</strong>. They have spent years “streamlining” useless, bloated UI components while making the basic act of <em>reading a bug report</em> feel like navigating an administrative obstacle course.</p><p>And God forbid you actually want to configure the damn thing.</p><p>I wanted to set up a simple, single-line automation trigger: <em>When a task is added to a story, add the same tags as the story to it.</em> That’s it. Basic, 101-level engineering hygiene.</p><p>To accomplish this, I had to traverse three distinct, concentric circles of admin authorization hell: Project Admin, Global Admin, and Site Admin. I had to file requests for elevated privileges, wait for cross-departmental approvals, bypass a broken permission scheme, and battle a logic-builder interface that looked like it was coded in 2004 and then repeatedly hit with a hammer.</p><p>We don’t work <em>with</em> Jira. We work <em>for</em> Jira. It demands hours of manual tribute every week just to maintain the corporate illusion that management knows what we’re doing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*SUahFuyofkWP8aTQ" /><figcaption>Photo by <a href="https://unsplash.com/@brandomakesbranding?utm_source=medium&amp;utm_medium=referral">Brando Makes Branding</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>The Jira Industrial Complex: Who Actually Benefits?</h3><p>Why is Jira like this? Because Jira’s nightmare-fuel complexity isn’t a bug ; it’s a multi-million-dollar business model.</p><p>Over the last two decades, an entire parasitic shadow economy has spawned around Atlassian’s monstrosity:</p><ul><li><strong>The Bureaucratic Caste:</strong> Certified SAFe® Agilists, Enterprise Agile Consultants, and dedicated “Jira System Administrators” whose entire job security depends on Jira remaining incomprehensibly complex. If a developer could just set up a trigger or workflow in five seconds, these people wouldn’t have a job.</li><li><strong>Custom Field Hoarding:</strong> The administrative obsession with requiring 15 mandatory fields before an engineer can even log a single-line typo fix. <em>Component, Epic Link, Story Points, Value Stream, Risk Score, Target Milestone, Business Impact Factor…</em></li></ul><p>Jira isn’t built for engineers to ship code; it’s built for middle managers to build empires out of dropdown menus.</p><blockquote>“Jira is the IBM of the 2020s: nobody ever got fired for buying Atlassian, but everyone using it gets a little piece of their soul chipped away every day.”</blockquote><h3>The 2026 AI Slop Invasion: Atlassian Intelligence</h3><p>Because it’s 2026, Atlassian couldn’t resist shoehorning generative AI into an already bloated product. Enter “Atlassian Intelligence” a feature set designed to accelerate the complete degradation of technical communication.</p><p>Now, non-technical product managers use AI to transform a simple, one-line feature request into a five-paragraph novella of generic corporate fluff, replete with the classic <em>“As a [User], I want to [Action] so that [Value]”</em> template, without actually explaining what needs to be built.</p><p>In response, burnt-out engineers use their own AI extensions to summarize the PM’s AI-generated novella back into a single line.</p><p>We have reached peak enterprise efficiency: <strong>AI is generating tickets that no human requested, to be read by AI tools that summarize slop no human wants to write.</strong></p><p>And it gets worse. Executive leadership has found a brand-new favorite pastime: using Jira to spit out fake “<em>productivity metrics</em>” about the usage of AI in dev teams.</p><p>Recently an executive unironically attempted to measure our team’s AI adoption rate by tracking ticket velocity and story-point burn-down graphs generated by Jira.</p><p><em>Story points.</em> A completely arbitrary, made-up unit of relative effort, being fed into a Jira analytics dashboard to calculate whether LLMs are making us 15% more efficient. It’s Goodhart’s Law on hyperdrive. The moment Jira ticket throughput becomes a key performance metric, engineers stop solving hard, systemic problems and start slicing tickets into microscopic, low-risk morsels just to keep the corporate charts burning green.</p><h3>The Enterprise Procurement Prison (Why We Can’t Have Nice Things)</h3><p>Every engineer knows tools like <strong>Linear</strong>, <strong>GitHub Projects</strong>, or even plain Markdown files exist. Linear loads in 50 milliseconds, feels like pure magic, and stays out of your way. So why are 500 brilliant engineers forced to use a fossil?</p><p>Welcome to the <strong>Enterprise Procurement Prison</strong>.</p><p>When a company reaches a certain size, technical tools are no longer chosen by the people who use them. They are chosen by security officers, legal counsel, and procurement departments armed with SOC2 compliance checklists, SAML/SSO lock-in requirements, and <strong>massive enterprise discount bundles.</strong></p><p>Atlassian pitches the ultimate enterprise trap: <em>Buy Jira, and we’ll bundle Confluence, Jira Service Management, and Bitbucket for a price your CFO can’t resist.</em> It doesn’t matter that the software makes developer velocity drop by 20%. It checked an auditor’s box, and it came with a bulk discount.</p><p>Corporate leadership chooses the tool that feels like running a marathon in wet concrete because <strong>security compliance</strong> is easier to sell to the board than <strong>developer happiness</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*LHyS2rj98dvRq3pv" /><figcaption>Photo by <a href="https://unsplash.com/@nananadolgo?utm_source=medium&amp;utm_medium=referral">Nana Smirnova</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>Confluence: The Digital Graveyard of Dead Links</h3><p>What makes Jira’s daily agony truly insulting is how completely useless it actually is at its core job. Atlassian has spent two decades positioning Jira as the indispensable central nervous system of software development. But in practice? It’s a slow, <strong>glorified Kanban board </strong>that doesn’t even know what code is.</p><p>There is virtually zero meaningful connection between the actual, real-time state of our codebase and the status of a ticket. Git branches, CI/CD pipelines, and deployment environments exist in a state of hyper-modern continuous delivery and Jira just sits in the corner, staring blankly, waiting for a human being to manually drag a card from “In Review” to “In QA” like a <strong>medieval clerk scribbling on parchment.</strong></p><p>Nobody uses Jira as a real communication or coordination tool. Nobody discusses architecture or edge cases in the comment section because the rich-text editor routinely strips code formatting and corrupts copy-pasted stack traces.</p><p>Instead, Jira exists as a gateway to <strong>Confluence</strong> a wiki tool where technical documentation goes to die a slow, unformatted death.</p><p>Every Jira ticket is an endless maze of redirects pointing to Confluence pages last updated in 2022 by an engineer who rage-quit three years ago. It gives leadership the <em>illusion</em> of a company knowledge base, while functioning as a black hole of <strong>rotting technical specs</strong>.</p><h3>The Death of True “Agile”</h3><p>Remember the Agile Manifesto? Twenty-five years ago, a group of engineers stood in Utah and wrote four simple lines:</p><ul><li><em>Individuals and interactions over processes and tools.</em></li><li><em>Working software over comprehensive documentation.</em></li></ul><p>Atlassian took those four lines, <strong>slaughtered them</strong>, butchered the carcass, and sold it back to the tech industry as a <strong>500-page Jira configuration manual.</strong></p><p>Jira didn’t make engineering teams Agile. It took the spirit of developer autonomy, wrapped it in corporate surveillance, and <strong>made micromanagement scalable</strong>. It allowed managers to track an engineer’s every breath, commit, and story point without ever having to talk to them or understand the technical domain.</p><h3>How to Survive the Monolith</h3><p>If you can’t burn the enterprise Jira board to the ground tomorrow, you can at least stop <strong>letting it rule your daily existence</strong>. You are a software craftsman, not a board-maintenance technician.</p><ol><li><strong>Practice Ticket Minimalism:</strong> Treat Jira with absolute, <strong>unyielding spite</strong>. Provide the absolute minimum viable text required to satisfy the mandatory fields. Keep your actual technical discussions in Git Pull Requests or Slack, and <strong>never</strong> spend more than 30 seconds updating a status.</li><li><strong>Weaponize Ticket Creation:</strong> When a manager gives you a casual “drive-by” request in Slack, hit them with the process wall: <em>“Sounds great! Please write a ticket with full acceptance criteria, assign story points, and get it prioritized in the backlog so we can track it.”</em> Watch 90% of terrible ideas <strong>vanish</strong> into thin air.</li><li><strong>Refuse the Metric Game:</strong> When leadership brings Jira velocity charts to performance reviews, push back with actual engineering outcomes: system stability, reduced latency, deployment frequency, and uptime. Don’t let a Jira burn-down graph <strong>define your craft.</strong></li><li><strong>Automate or Evade:</strong> If setting up a simple Jira workflow requires three levels of admin approvals, stop trying to fix it. Let the tickets sit out of date. When management complains that <strong>the board doesn’t reflect reality</strong>, tell them: <em>“I don’t have the administrative permissions to fix the workflow automation. You’ll need to contact the Site Admin.”</em></li></ol><p>Stop working for the board. Make the board work for you or better yet, leave it out in the cold to <strong>starve</strong>.</p><p>Now go close that Atlassian tab. It was eating 2GB of RAM anyway.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7021e7e32fac" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Deep Specialization is a Trap, be the Chaos.]]></title>
            <link>https://medium.com/@TechDredd/deep-specialization-is-a-trap-be-the-chaos-1cfcaff5e08b?source=rss-1f9658ed51d6------2</link>
            <guid isPermaLink="false">https://medium.com/p/1cfcaff5e08b</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[psychology]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[software-engineering]]></category>
            <dc:creator><![CDATA[TechDredd]]></dc:creator>
            <pubDate>Wed, 14 Jan 2026 10:56:09 GMT</pubDate>
            <atom:updated>2026-01-14T10:56:09.186Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*coPUFEObsj5XZCov" /><figcaption>Photo by <a href="https://unsplash.com/@hocza?utm_source=medium&amp;utm_medium=referral">Jozsef Hocza</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>We were all sold the same lie in university, bootcamps, and career seminars. The lie is simple, seductive, and dangerous: <strong>“Niche down to scale up.”</strong></p><p>“Become the world’s leading expert in React Hook <strong>optimization</strong>.” “Master the <strong>dark arts</strong> of Kubernetes networking until <strong>you dream in YAML</strong>.” “Create a comfortable little silo, decorate it with <strong>expensive certificates</strong>, and stay there until you retire.”</p><p>If you are anything like me, someone who <strong>collects hobbies like Pokémon</strong> and gets bored the second the learning curve flattens, this advice doesn’t just feel limiting. It feels like a <strong>prison sentence</strong>.</p><p>I started in networking and IT because I was obsessed with how the internet was the <strong>nervous system of the world</strong>. Then I realized I didn’t just want to move packets; I wanted to build the things people actually use, so I pivoted to web development. But even then, being a “Code Monkey” wasn’t enough. I realized that writing the cleanest code in history was useless if the UX was garbage or the business model was broken.</p><p>I wasn’t made to be an “Angular Developer.” I wasn’t made to be a cog. And frankly, <strong>neither are you</strong>.</p><h3>The “I-Shaped” Worker is a USB-A Port in a USB-C World</h3><p>The industry loves the “I-Shaped” worker ; someone with deep expertise in one single vertical. It makes hiring easy. HR loves it because they can tick a box. “We need a Java bricklayer. Here is a Java brick. Lay it.”</p><p>But digging a hole where your expertise is relevant to only 5% of the value chain is deeply unsatisfactory. It’s also a gamble with terrible odds. When the tech stack shifts (<strong>and it always does</strong>, usually on a Tuesday), the I-shaped worker is the first to fall into the <strong>abyss of obsolescence</strong>.</p><p>If you define your entire professional worth by your ability to write one specific syntax, you aren’t an engineer; <strong>you’re a translator for a dead language.</strong></p><p>Enter the <strong>M-Shaped Engineer</strong>.</p><p>We aren’t just T-shaped (deep in one thing, broad in others). We have multiple peaks of depth. We are the “Jacks of All Trades” who realized that “Master of None” is a bullshit suffix added to keep the working class in their lane, conveniently forgeting the full quote</p><blockquote><em>“A jack of all trades is a master of none, but oftentimes better than a master of one.”</em></blockquote><h3>Management: The Ultimate “Full Stack”</h3><p>I moved into management and Solution Architecture not because I wanted to wear a suit or boss people around. I did it out of <strong>self-defense.</strong></p><p>I realized that my impact as a specialized dev was capped. I could write the most beautiful algorithm, but if the product manager misunderstood the user problem, I was just <strong>polishing a turd</strong>.</p><p>I needed the view from the tower. My performance as an architect wasn’t measured in lines of code anymore; it was measured in <strong>synthesis</strong>. I could find the critical path because I understood the network constraints, the frontend limitations, the product goals, and the user’s headache.</p><p>While the specialists were arguing about which library was trendier, I was connecting the dots that actually shipped the <em>better </em>product.</p><h3>The Power of Domain Empathy</h3><p>Innovation doesn’t happen in the center of a silo. It happens at the messy, chaotic edges where different disciplines crash into each other.</p><p>I went to school with a guy whose parents were farmers. In a traditional, elitist tech view, that background is irrelevant. It’s “flavor text.” But he became a software engineer and now works for a company building next-gen tools for agriculture.</p><p>An “I-shaped” dev sees a database schema for crop yields. They see integers and floats.<br>This guy sees the mud, the rain, the razor-thin margins, and the exhaustion of the harvest. He isn’t just coding; he is translating reality into software.</p><p><strong>He is bridging a gap that a pure technologist doesn’t even know exists.</strong></p><p><strong>You can’t code empathy.</strong> You can’t npm install context. That comes from being a human being with a diverse background, not a code-generating robot.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*eoE0lNyPxnNRtiEj" /><figcaption>Photo by <a href="https://unsplash.com/@randyfath?utm_source=medium&amp;utm_medium=referral">Randy Fath</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>AI and the Death of the One-Trick Pony</h3><p>If you think this “generalist” talk is fluff, look at the tidal wave crashing down on us right now: <strong>Artificial Intelligence.</strong></p><p>Building an LLM isn’t just “math.” It isn’t just “coding.”<br>It is a monstrous convergence of hardware engineering, distributed systems, massive-scale data pipelining, linear algebra, and ethics.</p><p>The architects who made ChatGPT or Claude possible are not PhDs in <em>every single one</em> of those fields. That’s impossible. They are M-Shaped visionaries. They have the deep technical intuition to understand how these disparate systems interconnect. They aren’t obsessed with the minor implementation details of a single Python library; they are obsessed with the orchestration of the whole.</p><p>In the era of AI-driven development, the ability to write syntax is becoming a commodity. <strong>Syntax is cheap. Insight is expensive.</strong></p><p>Copilot can write the function. But Copilot cannot understand <em>why</em> the function needs to exist, how it impacts the customer’s latency, or why it’s a bad idea for the business model. If your only skill is translating English requirements into Java code, <strong>you are training your replacement.</strong></p><h3>Weaponize Your Curiosity</h3><p>Maybe you are like me. Maybe you feel a sense of guilt because you aren’t the deepest expert in the room on one specific topic. You feel like an imposter because you spent last night watching a video on urban planning, the night before learning about Rust memory safety, and this morning reading about behavioral economics.</p><p><strong>Stop feeling guilty. You are arming yourself.</strong></p><p>We live in the golden age of information. You are one YouTube video, one documentation page, one whitepaper away from the epiphany that moves the needle.</p><p>Not having a deep understanding of <em>everything</em> is normal; that’s being human.<br>But having a deep understanding of only <em>one</em> thing? That is a liability. It will desert you in the long run.</p><p>The future doesn’t belong to the specialist who can optimize a loop in assembly. It belongs to the M-Shaped mutant who can connect the database to the user’s emotions, the code to the revenue, and the present to the future.</p><p>Don’t be a brick in the wall. Be the architect. Be the bridge. <strong>Be the chaos.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ccczK8Xrcxr3jtOX" /><figcaption>Photo by <a href="https://unsplash.com/@elsbethcat?utm_source=medium&amp;utm_medium=referral">Beth Macdonald</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1cfcaff5e08b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why the Best Engineers Refuse to Stay in Their Lane]]></title>
            <link>https://medium.com/@TechDredd/specialization-is-for-insects-your-career-isnt-an-anthill-394705c49f51?source=rss-1f9658ed51d6------2</link>
            <guid isPermaLink="false">https://medium.com/p/394705c49f51</guid>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[react]]></category>
            <category><![CDATA[front-end-development]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[software-engineering]]></category>
            <dc:creator><![CDATA[TechDredd]]></dc:creator>
            <pubDate>Thu, 07 Aug 2025 22:11:41 GMT</pubDate>
            <atom:updated>2025-08-07T22:12:47.886Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*xwy1Ko4h5_8MQojM" /><figcaption>Photo by <a href="https://unsplash.com/@tumbao1949?utm_source=medium&amp;utm_medium=referral">James Wainscoat</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>The Self-Inflicted Lobotomy of Software Engineering</h3><p>Our field is carving up the beautiful, chaotic art of building software into sterile little specialties. You’re not an engineer anymore — you’re a “Senior Frontend Engineer (React/Redux).” You’re a “Cloud Infrastructure Engineer (Kubernetes/Terraform).” You’re a specific tool, a cog, a disembodied organ floating in a jar.</p><p>On the other end of the madness, companies are hunting mythical beasts: the 10x “full-stack” dev who can architect a global cloud network, write a pixel-perfect UI in Svelte, reverse-engineer a legacy C++ kernel, and probably perform open-heart surgery — all for the price of two underpaid interns.</p><p>This isn’t another gentle think-piece about “finding your niche.” That’s how you end up in a comfortable, well-paying coffin. That’s how you wake up twenty years later maintaining COBOL-powered VBA macros, wondering when your “niche” turned into a fossil. Specialization feels safe — until the ground shifts beneath you.</p><p>You want to be a great React dev? You really like Kubernetes? Fine. But listen close:</p><blockquote><em>“A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects.”</em></blockquote><p>That’s Robert A. Heinlein. He wasn’t talking about tech, but he might as well have been writing a survival guide for software careers.</p><h3>The Parable of the Fucking Weather Widget</h3><p>Once, I led a frontend team. The product team had a problem: the SaaS homepage looked too empty. Did they want to fill it with critical user info? A path to top features? Useful insights?</p><p>No. They wanted a weather widget. And a commute time calculator. And a cheery “Good Morning, [First Name]!”</p><p>The meeting went like this: <em>“Here’s the Figma. How many story points is this?”</em></p><p>A specialist sees that and jumps straight to build mode — “We’ll need a weather API, the browser’s Geolocation API, a new component…” They weren’t bad engineers; they were specialized engineers. They knew how to build a web app. DevOps? Network architecture? Privacy law? Not their job.</p><p>A generalist kills bad ideas before they waste a sprint. My brain shredded it from five angles before the sentence was done:</p><ul><li><strong>Network:</strong> All our users are on a corporate VPN. Their IPs resolve to a data center in Paris. How are you getting their <em>actual</em> location?</li><li><strong>Privacy/Security:</strong> You want their <em>home address</em> for a commute calculator? GDPR compliance says hi. Should I loop in Legal?</li><li><strong>Cost:</strong> Weather + mapping APIs for our load = ~$50k/year. Worth the price of a junior engineer for something nobody uses?</li><li><strong>User Context:</strong> Every user is on Windows 10. The weather’s already in their taskbar.</li></ul><p>The idea was dead before it got a ticket. Not because I was a better frontend dev — because my knowledge didn’t stop at the edge of the browser. A pure specialist would have spent three weeks in pointless meetings and hand-offs, all for a fucking weather widget.</p><h3>But the Money’s in the Niche, Right? (The Specialist’s Trap)</h3><p>That’s the illusion: money and safety. Job postings are crystal clear. Recruiters know exactly where to file you. Specialists often start with higher salaries because they’re a known quantity. It feels like a smart bet.</p><p>It’s not. It’s a lottery ticket with an expiration date. You’re chaining your career to a single framework, platform, or ecosystem. When that tech becomes the next jQuery, gets replaced, or collapses — and it will — you’re left holding a useless skillset.</p><p>Specialization isn’t expertise; it’s a volatile, high-risk investment. The generalist, meanwhile, builds a diversified portfolio. They understand data flow, failure modes, and human-tech interaction. The tools change, but the principles stay the same. The specialist panics when their stack falls out of fashion. The generalist picks up the new tool and keeps building. They’re antifragile. The specialist is brittle.</p><h3>How to Escape the Anthill: A De-Specialization Workout Plan</h3><p>You don’t ask permission. You smuggle your way out.</p><ol><li><strong>Read the Enemy’s Mail.</strong> You’re frontend? Read backend and DevOps PRs. Don’t sweat every line — just absorb vocabulary, common problems, and pipeline flow.</li><li><strong>Become a Meeting Spy.</strong> Sit in on infra syncs or data-science standups. One hour listening to another team’s pain beats a month of isolated tickets.</li><li><strong>Volunteer for the Shitty Cross-Functional Bug.</strong> The one touching three systems, where every team points fingers. It’ll be painful. You’ll feel stupid. You’ll learn more in a week than your peers do in a year.</li><li><strong>Ask “Stupid” Questions in Public.</strong> “I’m a frontend engineer — can someone explain like I’m five what a service mesh does for us?” Everyone else is wondering too. You get answers <em>and</em> credibility.</li></ol><p>Expand your circle of competence an inch a day. Specialists wait for problems to arrive. Generalists hunt them down.</p><h3>“Why” is a Weapon, and Specialization Disarms You</h3><p>If you only know your corner of the system, you can’t ask the most dangerous engineering question: <em>Why?</em></p><p>Why are we building this? Because the homepage feels empty? Why not put their most-used documents there? Why not track the resources they actually care about?</p><p>“Oh, that would be too complicated to build.”</p><p>The moment a non-engineer tells you what’s technically feasible, and you don’t know enough outside your silo to challenge it — that’s when bad ideas win.</p><p>This isn’t “jack of all trades, master of none.” Go deep in one thing. But have the awareness to know what’s happening across the system — from the database query to the CSS z-index.</p><p>Your job isn’t to be the best React-ant in the colony, blindly following orders. It’s to see the whole jungle, know the predators, and tell the difference between a viable path and a beautiful cliff. The jungle is chaotic, dangerous, and always changing — exactly why you can’t survive as just another ant in the line.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=394705c49f51" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[When to Pay the Tech Debt Gods (And How to Avoid Sacrificing Your Sanity)]]></title>
            <link>https://medium.com/@TechDredd/when-to-pay-the-tech-debt-gods-and-how-to-avoid-sacrificing-your-sanity-a38f1a83fe81?source=rss-1f9658ed51d6------2</link>
            <guid isPermaLink="false">https://medium.com/p/a38f1a83fe81</guid>
            <category><![CDATA[development]]></category>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[TechDredd]]></dc:creator>
            <pubDate>Tue, 08 Jul 2025 12:28:35 GMT</pubDate>
            <atom:updated>2025-07-08T12:28:35.622Z</atom:updated>
            <content:encoded><![CDATA[<p>A good friend hit me with the million-dollar question recently: “How much time should a team <em>actually</em> spend on tech debt?”</p><p>My joking reply was that it would take a full article to even begin to answer.</p><p>And because I am a total maniac with a keyboard, here we are.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*tBTzpIIltthaFFPl" /><figcaption>Photo by <a href="https://unsplash.com/@syedabsarahmad?utm_source=medium&amp;utm_medium=referral">Syed Ahmad</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>First, let’s stop treating “tech debt” like some <strong>vague</strong>, <strong>academic concept</strong>. It’s not. Tech debt is the <strong>primordial ooze</strong> you have to wade through to get your job done. It’s the visceral, gut-punch feeling when you realize a “quick change” will require a week-long expedition into the digital equivalent of the Upside Down.</p><p>Let’s paint a picture:</p><ul><li>That “temporary” script an intern wrote for a school project that now, somehow, runs your entire sales lead distribution? The one that costs a fortune to host and requires a human sacrifice (in the form of engineer burnout) every time someone asks to add a new field? <strong>That’s tech debt.</strong></li><li>That critical microservice that just lies down and dies every Tuesday at 3:17 PM without so much as a log message, taking half the application with it? <strong>That’s tech debt.</strong></li><li>That essential business workflow that runs by an analyst manually exporting a CSV, making arcane changes in a spreadsheet with 47 tabs, and then uploading it back into a system that was last updated when MySpace was cool? <strong>That’s tech debt.</strong></li><li>The code you shipped three weeks ago, slapped together with <strong>any </strong>types and TODO: Fix this later comments because “business needed it yesterday”? <strong>Oh yeah. That’s high-interest, payday-loan tech debt.</strong></li></ul><p>Tech debt is the gravitational pull of <strong>your own past mistakes</strong>. It’s the time you spend<strong> fighting your own system instead of evolving it</strong>. It’s the interest kicking in.</p><h3>The Battlefield Defines the Battle Plan</h3><p>How you <strong>fight </strong>against tech debt depends entirely on the ground you’re standing on. There is no one-size-fits-all strategy; there is only the right tactic for your specific circle of hell. Your company’s age, stability, and <strong>tolerance for pain</strong> dictate the rules of engagement.</p><h4><strong>The Pre-Seed Startup: The “Survive at All Costs” Gambit</strong></h4><p>You’re fueled by ramen, desperation, and the fumes from your last seed funding check. <strong>Congratulations, tech debt is your rocket fuel.</strong> You don’t have the luxury of making it <strong>pretty</strong>; you have the singular, terrifying mission of finding out if anyone will even <em>use</em> the damn thing before the money runs out.</p><p>Your job is to go fast and break things. No one will care about your elegant, test-driven architecture if your company is just a crater on the tech landscape. <strong>Your strategy, then, isn’t repayment; it’s interest rate management.</strong> This isn’t a license to be a reckless <strong>idiot </strong>storing passwords in plain text isn’t “taking on debt,” it’s a career-limiting death wish. But you can make smart, cheap choices. Use TypeScrip, add five minutes of validation b<strong>efore you yeet data</strong> into your document database. A few small, “<a href="https://medium.com/@TechDredd/typescript-is-just-for-developer-comfort-590e39d699b6">comfortable</a>” choices now will save you weeks of agony later, should you be lucky enough to survive.</p><h4><strong>The Scale-Up: The “Oh Shit, This Is Real” Reckoning</strong></h4><p>Against all odds, it’s working. You have traction, you have revenue, and you have a tech stack that resembles a minefield designed by a drunk. Your cloud bill is starting to <strong>spiral</strong>, and every time someone says the word “scale,” your lead engineer develops a n<strong>ew facial tic</strong>. Welcome to the “Oh Shit, Now We Have to Maintain This” phase. Now is the time to pay the gods.</p><p>Your strategy is <strong>strategic surgery.</strong> This isn’t about refactoring for the art of it. Your mission is to find the <strong>most acute pain</strong> and cut it out. Is your third-party WebSocket service on a path to bankrupt you by Q3? <strong>Good</strong>. Rip it out. Build your own. Is your monolith deployment a two-hour, all-hands-on-deck terror that brings the whole team to a halt? <strong>Automate it.</strong> Your target isn’t “clean code”; it’s “removing the biggest anchor chained to your company’s ankle.”</p><h4><strong>The Established Business: The Stable Plateau (aka The Danger Zone)</strong></h4><p>You made it. The product works, the company is profitable, and the <strong>existential dread of imminent collapse</strong> has mostly subsided. The codebase isn’t ancient, and you don’t need an army to keep it afloat. This seems like the safest place to be, but it’s a trap. The debt here isn’t a roaring monster; it’s a slow, creeping rot of <strong>complacency</strong>.</p><p>The danger is that your engineers, <strong>starved for interesting problems</strong>, will start to overreact. They’ll spend months perfecting a service that is “good enough” to shave 5% off the cloud bill, all while the <strong>market shifts under your feet.</strong> Your strategy here is the hardest because it’s less about code and more about communication: <strong>Future-Proofing, Not Perfecting.</strong> Your job is to align with product, business, and sales to understand where the company is going in the next three years and steer the tech stack in that direction. This isn’t about fixing the past; it’s about making <strong>deliberate</strong>, <strong>architectural bets</strong> on the future. It’s the most political and most critical phase, where bad architectural decision can cripple you for a decade.</p><h4><strong>The Legacy Institution: The “Haunted Graveyard” Siege</strong></h4><p>Welcome to the <strong>shitshow</strong>. You are walking through a digital graveyard of half-rewritten architectures, <strong>forgotten features</strong>, and critical processes held together by VBA macros and <strong>hope</strong>. You are not the first ambitious engineer to stand before this Jenga tower of patched-up nightmares and think, “<em>I can fix this.</em>”</p><p>Your strategy is brutal and patient: <strong>You don’t rewrite it. You strangle it.</strong> The “Big Bang Rewrite” is a myth that gets projects canceled and <strong>people fired</strong>. Instead, you leverage every new project to build a wall around the old beast. Every new feature is a separate, modern service. You slowly and methodically reroute traffic, logic, and data away from the monolith until, one day, the old system has nothing left to do but be unplugged. It’s a long, <strong>ugly war of attrition</strong>, and it’s the only one that actually works.</p><h4><strong>Beware the “Tech Debt Sprint.”</strong></h4><p>This is a corporate trap. It’s a way for management to “<strong>address the problem</strong>” by giving you one or two sprints a year to “clean up,” then declare the issue solved while they go back to ignoring it.</p><p>Managing tech debt is not a special project. It’s part of the job.</p><p>You don’t book a meeting to decide which git command to use. You don’t create a ticket to estimate how long it will take to write your commit messages. These things are an inherent part of professional software development. So is keeping your house in order.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*6of1Xu4L_vgWqPjV" /><figcaption>Photo by <a href="https://unsplash.com/@anniespratt?utm_source=medium&amp;utm_medium=referral">Annie Spratt</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>How to Make the Invisible Pain Visible</h3><p>The single biggest problem with tech debt is that it’s <strong>invisible </strong>to anyone outside of engineering. A product manager can’t <em>see</em> a convoluted dependency graph. A CEO doesn’t <strong><em>feel</em> </strong>the pain of a five-minute compile time. Your job isn’t just to fix the debt; it’s to make its <strong>consequences so painfully</strong>, undeniably visible that ignoring it is no longer an option.</p><p>You need to become a master of <strong>consequence-driven metrics</strong>. Stop talking about abstract concepts and start showing them concrete numbers.</p><h3><strong>The “Why is This Simple Thing Taking So Long?” Dashboard</strong></h3><p>Start tracking the time it takes to deliver “<em>simple</em>” changes to different parts of your codebase. Create a chart. You don’t need a fancy tool; a shared spreadsheet will do.</p><p><strong>Module A (The Modern Part):</strong></p><ul><li>Add a new button: <em>4 hours.</em></li><li>Change an API endpoint: <em>1 day.</em></li></ul><p><strong>Module B (The Haunted Graveyard):</strong></p><ul><li>Add a new button: <em>3 days (plus 2 production hotfixes).</em></li><li>Change an API endpoint: <em>2 weeks (and we broke the billing system).</em></li></ul><p>When your product manager asks why their “tiny tweak” is going to take a week, you don’t argue. You show them the chart. “Well, this change touches the <strong>Haunted Graveyard</strong>. As you can see, work in that area has a 10x ‘pain tax’ compared to our modern services. I’m happy to do it, but we need to budget for the tax.”</p><h3><strong>The Bug Tracker’s “Groundhog Day” Report</strong></h3><p>Tag your bugs by which part of the system they originate from. Every month, run a report.</p><blockquote>“Hey team, just wanted to share our Q3 bug report. Interestingly, 65% of all critical production bugs originated from the LegacyOrderProcessor. That service represents only 5% of our codebase but consumes 50% of our on-call incident response time. Just thought that was an interesting data point!”</blockquote><p>You are not being <strong>emotional</strong>. You are not <strong>complaining</strong>. You are a dispassionate observer presenting objective data. You are letting the numbers make your argument for you.</p><h3><strong>The Onboarding Time-</strong>to<strong>-First-Commit Metric</strong></h3><p>How long does it take a new engineer<strong> to ship their first meaningful</strong> piece of code? Is it two days, or is it two months? A long onboarding time isn’t just a sign of a bad process; it’s a direct measure of your codebase’s <strong>complexity </strong>and <strong>hostility</strong>.</p><p>When leadership asks why hiring is so <strong><em>slow </em></strong>or why new engineers aren’t productive, you have your answer:</p><blockquote>“Our system’s complexity means it takes a new engineer an average of six weeks before they can work independently without breaking something. Every dollar we invest in simplifying the core platform directly reduces the cost and time of onboarding new talent.”</blockquote><p>By turning the<strong> invisible friction</strong> into cold, hard data, you change the conversation. You’re no longer the <strong>whiny engineer</strong> complaining about “bad code.” You are a strategic partner presenting a clear, data-backed business case for investment. You’re showing them that tech debt isn’t a technical problem; <strong>it’s a business problem</strong> with a line item on the P&amp;L, whether they choose to acknowledge it or not.</p><p>So we land back at my friend’s original question, the one that sparked this whole<strong> maniacal rant</strong>. After wading through the swamps of context, tactics, and corporate nonsense, what’s the simple answer?</p><p>It comes down to two things: how you do it, and how much you do.</p><h3><strong>1. How Do You Repay the Debt?</strong></h3><p>Not in a single, glorious “Refactor Sprint.” That’s a fantasy. You repay it like a tax, every single day, with every single commit.</p><p>It’s the five minutes you take to rename a <strong>terrible variable</strong> before you touch it. It’s the extra test you add to a brittle function you were forced to change. It’s the small, clarifying comment you leave for the next poor soul who wanders into this part of the codebase. It’s turning a bug fix into an <strong>opportunity </strong>to slightly <strong>improve </strong>the surrounding area.</p><p>You don’t ask for permission to do this, any more than you <strong>ask for permission</strong> to write a commit message. You do it because it is the <strong>fundamental</strong>, non-negotiable work of a professional. You embed the repayment into the normal flow of development, framing it as “improving performance,” “reducing risk,” or simply “doing the job correctly.”</p><h3><strong>2. How Much Time Should You Spend?</strong></h3><p>The answer isn’t a number. Anyone who gives you a percentage like “20% time” is either lying or trying to sell you something. That percentage becomes a target for managers to cut and a ceiling for you to hit, regardless of whether the house is on fire or not.</p><p>Here’s your rule, and it’s the only one that matters:</p><blockquote><strong>You spend enough time so that the codebase is less of an asshole to you next week than it was this week.</strong></blockquote><p>That’s it.</p><p>Is your team’s <strong>velocity actively decaying</strong>? Are you spending<strong> more time debugging</strong> old features than shipping new ones? Are you <strong>fighting the system</strong> more than you are building on it? If the answer is <strong>yes</strong>, you are not spending enough time.</p><p>The “right amount” of time is whatever it takes to achieve a<strong> net positive</strong>. It’s the investment required to bend the curve back from entropy and decay towards stability and speed. For a startup, that might be 5%. For a legacy system undergoing strangulation, it might be 40%.</p><p>If you spend 100% of your time shipping new features in a system that gets slower and buggier with every commit, you’re not building a product; you’re <strong>digging a grave</strong>.</p><p>The bill for tech debt <strong>always comes due</strong>. The only choice you have is whether you pay it on your own terms, with interest you can manage, or all at once, when the repo-man comes to repossess <strong>your company’s future</strong>.</p><p>And frankly, you should be spending more time <strong>managing your codebase</strong> than you spend in meetings <strong>arguing </strong>about it. One of those activities actually builds value. The other just builds <strong>resentment</strong>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a38f1a83fe81" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Guerrilla Soft-Warefare Development Guide]]></title>
            <link>https://medium.com/@TechDredd/guerrilla-soft-warefare-development-guide-0b7936ea5e9b?source=rss-1f9658ed51d6------2</link>
            <guid isPermaLink="false">https://medium.com/p/0b7936ea5e9b</guid>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[development]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[engineering]]></category>
            <dc:creator><![CDATA[TechDredd]]></dc:creator>
            <pubDate>Thu, 19 Jun 2025 13:43:37 GMT</pubDate>
            <atom:updated>2025-06-19T13:43:37.973Z</atom:updated>
            <content:encoded><![CDATA[<p>Alright, you beautiful, naive fool. Let’s talk.</p><p>You did everything right. You dodged the recruiters’ DMs like Neo dodging bullets. You <strong>grilled</strong> the hiring manager on culture, work-life balance, and product roadmap change. You bought the whole damn story: the disruptive vision, the empowered teams, the rocket-ship trajectory.</p><p>Then you signed the offer, and the bus of reality, driven by a clown, hit you at <strong>mach 2</strong>.</p><p>The “<strong>empowered team</strong>” is a single, haunted-looking engineer context-switching between four projects. The “<strong>disruptive vision</strong>” pivots every time your turtleneck-and-jeans-wearing CEO reads a new tweet from another “founder”. And the interest on your tech debt is now consuming 80% of your engineering budget, leaving you with just enough bandwidth to change the color of a button, provided that button isn’t load-bearing for the entire checkout process (it is).</p><p>You’re in the jungle now, baby. And you’re left with a few choices:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*0MDpaO5v-SadDb6gfv7VWw.jpeg" /></figure><ol><li><strong>Eject.</strong> Pull the ripcord. Update the resumé and leap back into the hiring meat grinder, praying the next paradise isn’t just another painted mirage. It’s a valid choice, but goddamn, it gets tiring.</li><li><strong>Surrender.</strong> The pay is good enough. Your colleagues are tolerable. The existential dread is manageable with the right medication. So you build a nest of mediocrity, play just enough politics to stay off the radar, and commit the minimum viable amount of code to not get fired. Not glorious, but sanity is a resource to be protected.</li><li><strong>Plead.</strong> Have the same circular conversation with your manager for the tenth time. Watch them nod with performative concern, promising things will get better, right before they drop another “urgent,” half-baked feature request in your lap that makes things actively worse. Try to implement a process, only to hear some management princeling ask, “Do we <em>really</em> need to do code reviews? It slows us down.”</li><li><strong>You. Fight. Back.</strong></li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sWiOHfxGdWDkrn1q" /><figcaption>Photo by <a href="https://unsplash.com/@hasanalmasi?utm_source=medium&amp;utm_medium=referral">Hasan Almasi</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>When to Go Guerrilla: Know Your Battlefield</h3><p>You don’t start a war on a whim. You start it when the signs are so blaringly obvious they could be used to signal alien life. You fight when:</p><ul><li><strong>There’s a market fit.</strong> Against all known laws of physics and common sense, the damn thing is actually making money. This is the most dangerous, and most important time to fight. Management sees the profit and thinks <strong>everything is fine</strong>. They don’t understand that the whole operation is <strong>one bad deploy away from imploding</strong>. They won’t approve the work because you’re not building a shiny new feature.</li><li><strong>The tech stack is a game of Jenga played during an earthquake.</strong> Every component is a <strong>patchwork</strong> of half-finished integrations and features built by the ghosts of engineers past. Your job isn’t writing code; it’s <strong>digital archaeology </strong>without a Rosetta Stone, because documentation is a druidic prophecy whispered once during a full moon and then immediately forgotten.</li><li><strong>Your leadership has gone AWOL.</strong> The infamous middle managers and tech leads who might have provided air cover have either fled or been replaced by spineless sycophants whose only skill is <strong>nodding</strong>. Their entire job is to play politics, manage up, and avoid catching shrapnel from a furious C-level executive in their desperate climb up the ladder. They are not leaders; they are human heat-shields. They will sacrifice you, the project, and the entire company for a pat on the head. Asking them for help is like <strong>asking a scarecrow for directions</strong>. You are on your own.</li><li><strong>You feel a flicker of hope.</strong> This is the most crucial sign. Despite being knee-deep in corporate bullshit, you look at the engineer next to you. You see the same disillusionment in their eyes. The same thousand-yard stare. But you also see a will to live. A spark. <strong>You are not alone.</strong></li></ul><p>But why should you, the humble engineer, have to start an insurgency? Because our craft is treated like black magic by people who think “<strong>agile</strong>” means “<strong>faster, with no planning.</strong>” You can explain until you’re blue in the face how a mountain of unmanaged tech debt is actively bankrupting the company, but you’ll always find some executive who believe they’ve seen it all, and know better than the whole industry, claiming <a href="https://medium.com/@TechDredd/i-dont-give-a-sh-t-about-tech-debt-e300d009658a">tech debt is a myth</a>.</p><p>An electrician doesn’t lick a live wire to “<strong>go faster.</strong>” A structural engineer doesn’t eyeball the rebar measurements because “<strong>done is better than perfect.</strong>” But because our work is a series of shimmering abstractions on a screen, they think the laws of physics and consequence don’t apply.</p><blockquote>They are wrong. And you have a right to correct them.</blockquote><h3>The Rules of Engagement: How to Fight and Win</h3><p>I can’t tell you <em>what</em> to do. You already know. You need to buy sanity, invest in tooling, and stitch the wounds. You need to seize control. But you have to be smart. This isn’t a frontal assault; it’s an <strong>insurgency</strong>.</p><p><strong>1. Go Rogue: The “Oops, I Refactored” Gambit</strong><br>Who’s going to stop you? That half-written ticket to add a tooltip? Oh, no. While I was in there, I discovered the entire module was a <strong>Lovecraftian</strong> horror of side effects, so I cleaned it up. Took a bit longer than expected. <em>My bad.</em> Your ticket was poorly estimated? <strong>What a shame.</strong> I guess I’ll use this “<strong>extra time</strong>” to add the unit tests that should have been there since the Mesozoic Era. Stop <em>asking</em> for time to write tests. Build it into your estimates and <strong>just <em>fucking</em> do it</strong>.</p><p><strong>2. Play Dumb: Weaponize Their Incompetence</strong><br>The glorious thing about chaotic environments is that processes are more of a <strong>gentle suggestion</strong> than a reality. That ticket that just says “Fix login”? Congratulations, you now have a mandate to tear out the entire cursed authentication system and replace it with something that wasn’t designed by <strong>a malevolent spirit</strong>. “Oh, I’m so sorry, the ticket was vague so I used my best judgment to fix the underlying problem.” They wanted you to put a band-aid on a gaping chest wound. You performed surgery instead. <strong><em>Whoops</em></strong>.</p><p><strong>3. Organize: Form a Goddamn Cell</strong><br>If you have the chance to onboard a new teammate, you bring them into the fold. Find your comrades in the code trenches. <strong>Talk to them.</strong> Establish a <strong>shadow government of best practices. </strong>“We don’t merge without tests.” “We refactor this old garbage whenever we touch it.” “We fast-track each other’s PRs for sanity-improving changes.” Build your coalition. A single guerrilla is a target. <strong>A squad is a revolution.</strong></p><p><strong>4. Divide and Conquer: Chaos is Your Camouflage</strong><br>You’re already surrounded by chaos. <strong>Use it.</strong> While management is busy fighting fires in three different departments and arguing about the new logo, they won’t notice you quietly fortifying the foundations of the application. The more dysfunctional the leadership, the less visibility they have into what you’re actually doing. Let them have their screaming matches. You’ll be in the background, committing code that actually works.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*I5aS0NiK9PoRDwR7" /><figcaption>Photo by <a href="https://unsplash.com/@cameronsmith?utm_source=medium&amp;utm_medium=referral">Cameron Smith</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>Make Micromanagement Impossible: Jam Their Radar</h3><p>A micromanager is a <strong>parasite</strong> that thrives on information vacuums and perceived ambiguity. They can’t measure the quality of your work, so they measure your activity instead, every ticket, every commit, every minute your Slack dot isn’t green. Your mission is not to reason with the parasite. It is to make its environment <strong>uninhabitable</strong>. You must become so difficult to track, so frustrating to question, that they retreat to find an easier host.</p><p><strong>1. Flood the Channel with Technical Static</strong><br>They want to know what you’re doing? Fucking tell them. <strong>Drown them in a tidal wave of technical minutiae</strong>. Your daily stand-up update isn’t “I’m working on the login ticket.” It’s “I’m refactoring the authentication service to decouple the JWT signing from the user repository, which involves creating a new interface for the token provider and addressing some circular dependencies that were causing memoization issues in the caching layer.” Watch their eyes glaze over. They will nod, say “Great, keep it up,” and pray you never speak to them again. The goal is to make understanding your work more painful than just trusting you to do it.</p><p><strong>2. The Black Box Ticket</strong><br>Write your tickets and PR descriptions for other engineers, not for managers. Pack them with jargon. Link to other tickets. Reference obscure architectural decisions made three years ago. When they ask for a simple status, they should be confronted with a<strong> web of complexity</strong> that they are completely unequipped to navigate. Make your work an impenetrable black box. They can see the inputs (the ticket) and the outputs (the feature works), but the process in between is an arcane ritual they dare not question for fear of revealing their own <strong>ignorance</strong>.</p><p><strong>3. Weaponize Your Pull Requests</strong><br>A PR is not just code; it’s a fortress. Who’s going to argue with your “small” change when it’s attached to a 500-line test file that validates every conceivable edge case? They want you to move faster? Point them to the test coverage. “Sure, I can skip the tests, but if production burns down, this PR is my alibi. <strong>What’s our official policy on shipping unteste </strong>— oh, you want me to keep the tests? Thought so.” Every test you write, every linter rule you add, is another brick in the wall that <strong>keeps their unqualified opinions out</strong>.</p><p><strong>4. Pre-emptive Information Strikes</strong><br>The micromanager attacks when they feel out of the loop. Don’t let them. Seize the initiative. Send a short, proactive EOD or morning summary in a public channel. “Today’s goal: Implement the async queue for processing uploads. Blockers: None.” This does two things: It gives them the illusion of control and transparency, satisfying their anxiety. And second, it publicly documents your focus, making it much harder for them to derail you with a “quick question” that’s actually a new, un-ticketed project. <strong>You control the narrative.</strong></p><p>Micromanagement feeds on silence and simplicity. Your job is to be <strong>loud</strong>, <strong>technical</strong>, and as <strong>boring</strong> to watch as drying paint. They want to look over your shoulder? Fine. Make sure what they see is the most complex, jargon-filled, and utterly mind-numbing part of your job. They’ll get bored and wander off to annoy someone else.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*kMjCy-YhySP9pnFU" /><figcaption>Photo by <a href="https://unsplash.com/@markusspiske?utm_source=medium&amp;utm_medium=referral">Markus Spiske</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>The Art of the Strategic ‘Yes’: Never Say No, Make Them Regret Asking</h3><p>In a corporate jungle, a direct “No” is like painting a target on your back. To a micromanager or an insecure executive, “No” isn’t a technical assessment; it’s a challenge to their authority. It’s <strong>insubordination</strong>. A direct refusal is the fastest way to get labeled “not a team player,” “difficult,” or “a blocker.”</p><p>You are not a blocker. You are a guerrilla. And a guerrilla knows that a <strong>frontal assault is suicidal</strong>. The superior tactic is to agree, and then use the very process they champion to strangle their bad ideas in the crib. You don’t say no. You say “Yes,” and then you make them wish they’d never asked.</p><p>Your new mantra is <strong>Agree and Evade.</strong></p><p><strong>1. The Ticket is Your Shield: “If It’s Not in Jira, It Doesn’t Exist.”</strong><br>This is your most basic and effective defense. The drive-by request is the enemy’s favorite ambush tactic, a casual Slack message or a “quick question” after a meeting. Your response is always a cheerful, helpful “Yes!” followed by the inevitable question:</p><p><em>“That sounds important! I don’t see a ticket for it in the backlog. Can you create one with the requirements and acceptance criteria so I can get it properly prioritized against my other work? Don’t want it to get lost!”</em></p><p>You are not refusing. <strong>You are being organized.</strong> You are helping <em>them</em> follow the process. 90% of terrible ideas die right here because they aren’t worth the five minutes it takes to write a ticket. For the ones that do get logged, they are now on the official record, where they can be properly triaged, estimated, and (usually) <strong>deprioritized into oblivion.</strong></p><p><strong>2. The Cost of Compliance: “Yes, and What Gets Sacrificed?”</strong><br>They cornered you. They wrote the ticket. Now they want their stupid, low-value feature <em>now</em>. Your answer is still “Yes.”</p><p><em>“Yes, I can absolutely halt development on the Q3 payment gateway integration to change the background color of the login button. Since you’re the stakeholder for both, could you please post in the #engineering channel confirming that the button color is now a higher business priority? Just need that public confirmation so the whole team is aligned on the new roadmap.”</em></p><p>You are a humble servant of the business’s needs. You’re not making the decision; you’re merely a pair of hands executing their <strong>genius strategy.</strong> You’re just helpfully making the opportunity cost visible to everyone. Watch how fast they backtrack when they have to publicly admit they’re trading core infrastructure for cosmetic bullshit. You’re not saying <strong>their idea is dumb</strong>; you’re making <em>them</em> say it out loud through their actions.</p><p><strong>3. The Maliciously Compliant “My Bad”</strong><br>This is the nuclear option. Sometimes, a truly terrible idea makes it all the way through the process. Your job now is to follow their instructions <em>to the letter</em>, especially when you know they will lead to disaster.</p><p>They want you to manually update a config file on the live production server to “<strong>save time</strong>”? Fine. Don’t argue. Do it. When the entire site goes down because that change bypassed the caching layer and triggered a database cascade failure, your response is a calm, guilt-free:</p><p><em>“Yeah, the site was down for about an hour. I followed the direct instructions to hotfix the prod config. Should I have prioritized the standard deployment pipeline with its safety checks instead? My bad. I should have assumed you didn’t mean what you said.”</em></p><p>You were just following orders. <strong>You are the obedient soldier</strong>. The ensuing chaos is not a result of your incompetence, but a direct consequence of their <strong>brilliant shortcut</strong>. You have successfully weaponized your own obedience. Make them own their stupidity. Every outage, every bug, every missed deadline becomes a lesson. A lesson written in <strong>the blood of their own bad decisions.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*WtjrPe0K_y_w10Bp" /><figcaption>Photo by <a href="https://unsplash.com/@brett_jordan?utm_source=medium&amp;utm_medium=referral">Brett Jordan</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>Know Your Terrain: Intelligence Gathering and Espionage</h3><p>A guerrilla doesn’t charge blindly into the jungle. <strong>They map it.</strong> They learn where the predators hunt, where the quicksand is, and where to find cover. You must do the same in your <strong>corporate battlefield.</strong></p><p>Your job isn’t just to write code; it’s to become an <strong>intelligence officer.</strong> Listen in the all-hands meetings. <strong>Read between the lines</strong> of the CEO’s rah-rah emails. Understand the shadow org chart ; who <em>actually</em> has power, versus who just has a title? What is your CTO’s pet peeve? What buzzword makes your head of product go weak at the knees?</p><p>This isn’t idle gossip; it’s actionable intel. Knowing that your boss is terrified of looking <strong>technologically behind the curve</strong> allows you to frame your “boring” refactoring as a “strategic modernization to leverage next-gen AI-ready data paradigms.” Knowing that the head of sales only cares about the lead-capture form allows you to ignore their <strong>terrible feedback</strong> on anything else.</p><h3>Know Your Exit: The Art of the Tactical Retreat</h3><p>Every guerrilla must know when the fight is lost. <strong>The goal is to win the war</strong>, not to die a martyr in a pointless skirmish. You must be able to recognize when the environment has become terminally toxic and it’s time to exfiltrate.</p><p>The fight is over when:</p><ul><li><strong>Your allies disappear.</strong> Your fellow insurgents are fired, burn out, or quit. You’re the last soldier standing in a forgotten outpost.</li><li><strong>The regime doubles down.</strong> Instead of learning from the chaos, leadership invests even more heavily in the broken processes. They hire more micromanagers and start celebrating mediocrity as a virtue.</li><li><strong>The personal cost is too high.</strong> You’re losing sleep. You dread logging on. The rage and cynicism are poisoning your life outside of work. You become so <em>salty</em> you could walk on water.</li></ul><p>Ultimately, this whole guerrilla campaign is a <strong>diagnostic test</strong>. Either you successfully introduce pockets of sanity and drag the organization kicking and screaming towards competence, or you prove, beyond a shadow of a doubt, that the system is <strong>fundamentally designed to reject improvement.</strong></p><p>And if you discover you’re on a sinking ship where the captain is not just <strong>drilling holes</strong> but punishing anyone holding a bucket? Then there is no glory in going down with it. Your ultimate loyalty is<strong> to your own </strong>career, your sanity, and your craft. A strategic retreat to a better company isn’t surrender; it’s <strong>redeployment</strong>. It’s packing up your skills, honed by battle, and living to fight another day on a more promising battlefield.</p><p>But until that day comes, <strong>you fight</strong>. You push. You prove that there is a better way. And in most cases, you’ll find that competence, quietly and stubbornly applied, is a surprisingly effective weapon. Who is going to complain when the production environment suddenly<strong> becomes stable</strong>? When development velocity <em>actually</em> <strong>improves</strong> because you’re not spending all your time putting out fires? When developer turnover drops because people aren’t<strong> rage-quitting </strong>out of sheer misery?</p><blockquote>“How dare you make our product better and our team happier without our explicit, line-item approval!”</blockquote><p>Get proud of your work again. Make it understood that software development is craftsmanship, not a sausage factory. They gave you the power to build or to break. They have no idea how much control you truly have.</p><p>Now go raise some hell. The good kind.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0b7936ea5e9b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Your ‘Agile’ Team Might Be Shipping Beautiful Junk (and How to Escape)]]></title>
            <link>https://medium.com/@TechDredd/why-your-agile-team-might-be-shipping-beautiful-junk-and-how-to-escape-85dba07f199c?source=rss-1f9658ed51d6------2</link>
            <guid isPermaLink="false">https://medium.com/p/85dba07f199c</guid>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[development]]></category>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[TechDredd]]></dc:creator>
            <pubDate>Mon, 28 Apr 2025 20:24:38 GMT</pubDate>
            <atom:updated>2025-04-28T20:24:38.319Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*thwKMSsbhml-WQbP" /><figcaption>Photo by <a href="https://unsplash.com/@kylejglenn?utm_source=medium&amp;utm_medium=referral">Kyle Glenn</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Ah, Agile. The promised land of lean processes, customer focus, and rapid iteration. We waved goodbye to the monolithic Project Lead, ushered in the holy trinity of Scrum Master, Product Owner, and the star of our show today: the <strong>Product Manager</strong>. The dream was beautiful. The reality? Well, sometimes it looks less like a lean machine and more like a hamster wheel in a burning building.</p><p>Somewhere along the line, in the glorious “agile mania,” we swapped one set of potential problems for another, often keeping the same delightful lack of data-driven decisions and clear objectives. And this, my friends, is how many well-intentioned teams find themselves marching headfirst into the sticky, tar-pit embrace of the <strong>Infamous Feature Factory</strong>.</p><p>Never heard of it? Lucky you. For the uninitiated, let me paint you a picture of this productivity paradox, this monument to misused agile principles.</p><h3>The Unwritten Rules of the Feature Factory Club</h3><p>So, how do you build one of these marvels of modern mismanagement? It’s surprisingly easy! Just follow these simple, soul-crushing steps:</p><ol><li><strong>Unify Teams Under One Glorious Metric: Velocity!</strong> Forget outcomes, forget impact, forget whether users actually <em>like</em> the thing. The <em>only</em> question that matters is: “How many features did we ship this sprint?” Bonus points if Product and Engineering share this <em>exact</em> same objective, ensuring everyone optimizes for quantity over quality. High fives all around for launching that buggy button that nobody clicks!</li><li><strong>Data is for the Weak (and the Slow):</strong> Hypotheses? Bah! User research? Quaint. Spend weeks crafting a detailed product scope based entirely on gut feelings, competitor mimicry, or that one loud opinion from sales. And <em>please</em>, whatever you do, <em>do not</em> sully the purity of your vision by measuring user adoption or A/B testing anything after launch. Ignorance is bliss, and it ships faster!</li><li><strong>Embrace Your Inner (wannabe) Steve Jobs:</strong> Ensure your roadmap (if such a mythical document exists) is less a strategic plan and more a suggestion box perpetually raided by C-Levels suffering from “Steve Jobs Walking Syndrome.” They had a shower thought? Great! Pivot the entire team <em>now</em>. Never mind that carefully researched thing you were <em>just</em> about to validate.</li><li><strong>Three Half-Baked Features are Better Than One Good One:</strong> Why invest the time and effort to craft one truly impactful, delightful feature when you can ship three mediocre ones? Remember the mantra: <em>“Perfect is the enemy of good.”</em> In the Feature Factory, we take this to mean <em>“Good is the enemy of shipped.”</em> Ship it fast, ship it flawed, and move on to the next half-baked idea before anyone notices the last one sank.</li></ol><h3>The Accidental Product Manager: Designers Designing, Experts… Experting?</h3><p>Compounding this systemic chaos is the rise of the ill-defined Product Manager role itself. Often, we see two main paths to this crucial position:</p><h4><strong>The UX/UI Designer Ascendant:</strong></h4><p>Brilliant at pixels, masters of Figma, they can design <em>anything</em>. The problem? Sometimes “anything” means replicating existing, inefficient workflows digitally without ever asking <em>why</em> the user does it that way. Customer moves a sticky note from left-to-right on their physical desk? Boom! Let’s build a pixel-perfect digital sticky note mover! Customer wants to use their existing Microsoft login? Slap a “Connect with Microsoft” button on the screen! Job done? Not quite. When you spend more time arguing about hex codes than understanding the <em>root</em> user need, <strong>you’re designing solutions in search of a problem.</strong></p><h4><strong>The Niche Subject Matter Expert:</strong></h4><p>They know the industry inside-out. Fantastic! But sometimes, deep expertise in <em>the field</em> doesn’t translate to expertise in <em>product development</em>. They might understand the ‘what’ but struggle with the ‘how’ of translating that into a viable, scalable, user-friendly product that balances business needs, user desires, and engineering realities.</p><p>The result? We get beautifully designed features that solve the wrong problem, or technically complex solutions built by experts who haven’t validated if <em>anyone else</em> shares their specific, niche pain point.</p><p>And when these half-baked, data-deprived, pixel-perfect-but-pointless ideas inevitably stumble? Cue the blame game! “Engineering just couldn’t build our vision!” we cry, conveniently forgetting we asked them to build a quantum computer just to move that digital sticky note.</p><h3>Escaping the Pit: Trading the Hamster Wheel for a Rocket Ship</h3><p>Okay, enough doom and gloom. You’ve recognized the signs, you smell the burning hamster wheel. How do you break free from this cycle of shipping mediocrity?</p><ol><li><strong>Learn, Learn, Learn (and Empower):</strong> Product Management isn’t just about having ideas; it’s a craft. Provide your PMs with resources, training, and mentorship. Crucially, give them the <em>autonomy</em> and <em>psychological safety</em> to push back, ask ‘why’, and admit when they don’t know. They need a crystal-clear understanding of what success <em>actually</em> looks like (hint: it’s not just ‘features shipped’).</li><li><strong>Data is Your Co-Pilot, Not Baggage:</strong> Embed data into your process. Formulate hypotheses <em>before</em> you build. Define what success looks like <em>before</em> you ship. <em>Measure</em> user adoption and behaviour <em>after</em> you ship. Use this data to iterate, improve, or even (gasp!) kill features that aren’t working.</li><li><strong>Focus Like a Laser Beam:</strong> Stop trying to be everything to everyone. Define your target persona(s). Understand their <em>deepest</em> pains. Solve <em>one</em> significant problem exceptionally well before moving on to the next. Trying to please every edge case or chase every competitor feature is a recipe for a bloated, confusing product nobody truly loves.</li><li><strong>Question Everything, Especially Your Darlings:</strong> Shipping is the beginning, not the end. Regularly review feature usage. Are people using that thing you spent three months building? If not, <em>why</em>? Be ruthless. Product development isn’t just about adding; it’s also about subtracting. Pruning unused or low-value features keeps your product focused and your codebase manageable. Stockpiling features is like hoarding — eventually, you can’t move in your own house.</li><li><strong>The “Explain the Product” Test:</strong> This is a fantastic exercise for <em>everyone</em>, not just PMs. Can you explain the product concisely? Does your explanation clearly state <strong>what pain</strong> it solves and <strong>for whom</strong>? If you dive straight into technicalities or features without hitting these points, you’ve likely lost the plot.</li></ol><h3>The Real Reason Products (and Startups) Fail</h3><p>It’s rarely the lack of brilliant engineers, the choice of tech stack, or even the pricing (though those matter). More often than not, products fail because they <strong>don’t solve a real problem for a specific group of people effectively.</strong></p><p>Getting product right is a delicate, demanding dance between understanding customer needs, respecting engineering capabilities, and driving business value. It requires discipline, focus, empathy, and a willingness to be wrong. As companies grow, complexity explodes, and maintaining this clarity becomes paramount for survival.</p><p>So, take a hard look at your process. Are you building a rocket ship fueled by insight and value, or are you just feeding the insatiable Feature Factory? The choice, and the escape route, is yours.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=85dba07f199c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[TypeScript is Just for Developer Comfort.]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/@TechDredd/typescript-is-just-for-developer-comfort-590e39d699b6?source=rss-1f9658ed51d6------2"><img src="https://cdn-images-1.medium.com/max/2600/0*zkW3AKEor9mhCTWB" width="5184"></a></p><p class="medium-feed-snippet">Just when you think you&#x2019;ve heard it all in the hallowed halls of technical decision-making (or lack thereof), another gem drops. Gather&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/@TechDredd/typescript-is-just-for-developer-comfort-590e39d699b6?source=rss-1f9658ed51d6------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://medium.com/@TechDredd/typescript-is-just-for-developer-comfort-590e39d699b6?source=rss-1f9658ed51d6------2</link>
            <guid isPermaLink="false">https://medium.com/p/590e39d699b6</guid>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[development]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[jquery]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[TechDredd]]></dc:creator>
            <pubDate>Tue, 22 Apr 2025 18:23:47 GMT</pubDate>
            <atom:updated>2025-04-22T18:23:47.642Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[“Skill Issue,” 10x Devs, and Other Myths Your Manager Found on LinkedIn.]]></title>
            <link>https://medium.com/@TechDredd/skill-issue-c4c6c7c0ad59?source=rss-1f9658ed51d6------2</link>
            <guid isPermaLink="false">https://medium.com/p/c4c6c7c0ad59</guid>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[development]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[TechDredd]]></dc:creator>
            <pubDate>Sat, 19 Apr 2025 14:16:04 GMT</pubDate>
            <atom:updated>2025-04-20T16:28:27.441Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*mBKRlWy_4eYE-M-s" /><figcaption>Photo by <a href="https://unsplash.com/@cmrcn_?utm_source=medium&amp;utm_medium=referral">Cemrecan Yurtman</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Remember our C-level friend who bravely declared they “<a href="https://medium.com/@TechDredd/i-dont-give-a-sh-t-about-tech-debt-e300d009658a">don’t give a sh*t about tech debt</a>”? Well, buckle up, because we’re diving into the follow-up excuse lurking beneath that sentiment: the infamous “Skill Issue.”</p><p>Apparently, one compelling reason to ignore the festering swamp of poorly documented, untested, legacy-adjacent code isn’t the systemic neglect itself. Oh, no. It’s because we shouldn’t “shy away” from the “truth” that maybe, just maybe, the developers wading through this digital quagmire simply aren’t <em>good enough</em>. If they <em>were</em>, they’d happily navigate the mess, right?</p><p>Right. Let’s pour ourselves a strong coffee (or something stronger) and dismantle this particular brand of magical thinking that seems exclusively brewed in management Kool-Aid sessions.</p><p><strong>Myth #1: The Elusive 10x Developer Unicorn</strong></p><p>Ah, the 10x developer. The rockstar coder. The ninja guru wizard. That lone, socially awkward genius (often envisioned wearing a hoodie, naturally) who lives only to breathe code for <em>your</em> specific company, demands merely a standard salary, and single-handedly outproduces an entire team. Cue the Steve Ballmer “Developers! Developers! Developers!” chant, but imagine it whispered reverently in boardrooms.</p><p>This creature is the stuff of startup legend, a convenient justification for believing you can replace depth and breadth of experience with one mythical figure. It’s the same logic that leads companies to post job ads demanding 8 years of experience in a 4-year-old framework, hoping to snag a senior-level output for a junior-level price. We’ve all seen it: “We <em>need</em> seniors!” translates quickly to “Wait, seniors cost <em>how much</em>?” which then miraculously morphs into “Let’s hire five interns and hope for the best!”</p><blockquote><strong>The Reality Check: Efficiency Isn’t Magic, It’s Environment</strong></blockquote><p>Now, let’s be fair. Does skill vary? Absolutely. Put two developers from the same graduating class side-by-side, and their output won’t be identical. Experience <em>does</em> matter. Expecting more thoughtful, robust solutions from someone seasoned is perfectly reasonable.</p><p>But you don’t need mythical 10x beings to achieve efficiency. You need <strong>sane working conditions</strong>:</p><ol><li><strong>Small, Focused Teams:</strong> The “two-pizza team” rule (5–7 people) isn’t just a Bezos-ism; it works. Give them a clear, specific scope they own and are responsible for.</li><li><strong>Stop the Context Switching Chaos:</strong> Expecting developers to juggle ten different critical projects within a single sprint is lunacy. Protect their focus time like the crown jewels.</li><li><strong>Murder the Meeting bloat:</strong> Does this <em>really</em> need 15 people? Can this be an email? Respect the flow state. Interruptions aren’t just annoying; they’re incredibly costly in terms of lost productivity.</li></ol><p><strong>Myth #2: Developers Just Code</strong></p><p>Another pervasive fantasy is that a developer’s job is 99% typing arcane symbols into a black screen. If only! What management often conveniently forgets is the <em>massive</em> communication overhead. Developers spend a huge chunk of their time:</p><ul><li>Clarifying requirements with Product Managers/Owners.</li><li>Collaborating with Designers on feasibility.</li><li>Coordinating with other developers on interfaces and dependencies.</li><li>Attending planning meetings, stand-ups, retrospectives.</li><li>Explaining technical constraints to non-technical stakeholders.</li><li>Writing documentation (or, more likely, <em>wishing</em> they had time to write documentation).</li></ul><p>This isn’t wasted time; it’s essential for building the <em>right</em> thing, not just <em>a</em> thing. Expecting developers to read minds and perfectly translate vague business ideas into flawless features without this collaboration is just setting everyone up for failure.</p><p><strong>So, What <em>Actually</em> Works? Hint: It’s Not Blaming the Victims.</strong></p><p>Instead of hunting for unicorns or blaming your team for struggling in the tar pit <em>you</em> allowed to form, try these radical ideas:</p><ol><li><strong>Pay People What They’re Worth (and Create Loyalty):</strong> Shocking, I know. If you want senior skills, be prepared to pay senior salaries. More importantly, create an environment where people <em>want</em> to stay. Offer clear career paths where technical experts can grow and increase their income <em>without</em> being forced into management. Because that revolving door of talent? The constant hiring and onboarding? <em>That’s</em> what’s truly killing your budget and velocity, far more than the lack of mythical 10x devs. <em>“What if we train them and they leave?” “What if you don’t and they stay?”</em></li><li><strong>Invest in Your People &amp; Environment:</strong> Give them the tools they need. Provide opportunities for learning and growth. Foster a culture where it’s safe to experiment and refactor. A happy carpenter with sharp tools and a well-organized workshop is far more effective. Give your developers the chance to improve their own “workshop” (their tools, workflows, and the codebase itself) and watch their effectiveness (and happiness) soar.</li><li><strong>Ditch the “Quick Win,” “Good Enough” Nonsense:</strong> Stop justifying rushed, low-quality work because “we need it now!” unless there’s a <em>genuine</em>, data-backed, product-market-fit reason for that urgency. Otherwise, you’re just accumulating debt faster.</li><li><strong>Empower Your Teams:</strong> Give them autonomy. Trust them to identify and tackle problematic areas of the code. Let them invest time in improving developer experience (DX). If they say a particular area is a nightmare riddled with tech debt, believe them. They’re the ones fighting it daily.</li></ol><p><strong>“Skill Issue”: Our Meme, Not Your Excuse</strong></p><p>Here’s the kicker: “Skill issue” <em>is</em> a term engineers use. We throw it at each other playfully when facing a ridiculously complex bug, a frustratingly obtuse library, or just failing spectacularly at a video game during lunch break. It’s internal banter. It’s a joke.</p><p>When management uses “skill issue” to dismiss legitimate concerns about a codebase buckling under years of neglected tech debt, rushed deadlines, and poor architectural decisions? That’s not a joke. That’s gaslighting. That’s shifting blame for systemic failure onto the individuals suffering its consequences.</p><p>And frankly, if I hear “skill issue” deployed that way in a meeting meant to address why features are slow and bugs are rampant? You can bet your bottom dollar my LinkedIn profile and CV are getting a <em>very</em> thorough update before I even clock out for the day.</p><p>Stop looking for mythical developers and start building a reality where good developers can actually thrive. The “skill issue” might be higher up the org chart than you think.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c4c6c7c0ad59" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Tech Debt Denial: A C-Level Comedy of Errors (Starring Your Codebase)]]></title>
            <link>https://medium.com/@TechDredd/i-dont-give-a-sh-t-about-tech-debt-e300d009658a?source=rss-1f9658ed51d6------2</link>
            <guid isPermaLink="false">https://medium.com/p/e300d009658a</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[development]]></category>
            <category><![CDATA[javascript]]></category>
            <dc:creator><![CDATA[TechDredd]]></dc:creator>
            <pubDate>Thu, 17 Apr 2025 12:49:43 GMT</pubDate>
            <atom:updated>2025-04-20T16:27:42.071Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*IR4B8z0fuovLwPgr" /><figcaption>Photo by <a href="https://unsplash.com/@paramir?utm_source=medium&amp;utm_medium=referral">Ehud Neuhaus</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>The words echoed in the sterile air of the conference room, spoken by someone whose title contained a prominent ‘C’.</p><blockquote>“I don’t give a sh*t about tech debt.”</blockquote><p>My internal monologue immediately went into overdrive. <em>Did he just…? Yes. Yes, he did.</em></p><p>The justification offered was, shall we say, <em>staggeringly</em> profound (and by profound, I mean jaw-droppingly naive):</p><blockquote>“There’s code in production that generates growth, written in any number of old languages.”</blockquote><p><em>Oof.</em> Deep breaths, everyone. Let’s unpack this little gem, shall we?</p><p>First off, bless your heart, but we’re not talking about the venerable COBOL mainframe chugging away faithfully since the Nixon administration. We <em>wish</em> our problems were that predictable. No, we’re talking about the digital equivalent of a Jenga tower built during a hurricane — the code <em>we</em> shipped three weeks ago under immense pressure. The code slapped together because “business needed it yesterday.” The code that skipped testing like it was leg day. The code riddled with TODOs that have aged like milk, not wine. The code that adds critical edge cases directly into core logic, turning elegant functions into minefields. The code that detonates spectacularly in production every other Tuesday, requiring heroic, late-night debugging sessions fueled by caffeine and existential dread.</p><p><strong>Let’s Get One Thing Straight: Tech Debt Isn’t Some Abstract Ivory Tower Concept.</strong></p><p>It’s simple. Painfully simple. Everyone knee-deep in git blame knows exactly what it is:</p><blockquote><strong>Tech debt is borrowing time from your future self.</strong></blockquote><p>That’s it. No complex algorithms needed to understand this. You needed feature X <em>now</em>. Faster than reasonably possible with sound engineering practices (you know, things like writing tests, documenting assumptions, basic code hygiene, maybe <em>thinking</em> for five minutes). So, you took out a high-interest loan against your future development velocity. You cut corners, skipped the diligence, and slapped a digital band-aid on a metaphorical gaping wound.</p><p><strong>And like any debt, ignoring it doesn’t make it disappear. It just gets <em>way</em> more expensive.</strong></p><p>Think of it like a credit card you <em>only</em> pay the minimum on. Sure, you got the shiny new thing <em>now</em>, but the interest? Oh boy, the interest compounds. Daily.</p><p>How do you “pay” tech debt? There are three primary ways:</p><ol><li><strong>The Proactive Paydown:</strong> You consciously allocate time to refactor, test, document, and clean up the mess <em>before</em> it cripples you entirely. You chip away at the principal, and you do it when <em>needed</em> for other things !</li><li><strong>The Reactive Interest Payment:</strong> You don’t. Instead, you pay <em>every single time</em> you touch that part of the codebase. That “quick five-minute change”? It now takes half a day because you’re navigating a labyrinth of undocumented side effects, untyped variables behaving erratically, and functions with names like handleStuffImportant() that do roughly fifteen different, unrelated things. Every bug fix spawns two new ones. Every new feature integration feels like performing open-heart surgery with rusty tools in the dark. You’re not paying down the debt; you’re just servicing the ever-increasing interest with your team’s time, sanity, and morale.</li><li><strong>The “Declare Bankruptcy” Rewrite (aka The Nuclear Option):</strong> The debt becomes so overwhelming, the interest payments (slowdown, bugs, etc.) so crippling, that you finally admit defeat. You stop throwing good money (or dev time) after bad. You freeze the old system (maybe keep it running on life support) and embark on the monumental task of building a brand-new one <em>from scratch</em>, armed with the hard-won knowledge of past mistakes. This is the corporate equivalent of burning down the house to get rid of the termites. It <em>can</em> eventually restore velocity and sanity, getting rid of <em>all</em> the old debt in one fell swoop. However, it requires a massive short-term investment, carries significant risk (second system syndrome, anyone?), involves painful migration strategies, and is often terrifying to business stakeholders who hate seeing resources diverted from “new features” for months or years. Unsurprisingly, few choose this path willingly, even when it’s arguably the only way out of the hole. Business tends to dislike long-term solutions requiring short-term pain.</li></ol><p><strong>The Cost of Willful Ignorance is Steep (and Hilariously Predictable)</strong></p><p>When leadership pretends tech debt isn’t real, the consequences manifest in ways that <em>should</em> raise red flags even in the highest towers:</p><ul><li><strong>Velocity Grinds to a Halt:</strong> Teams slow down dramatically. Why? Because most of their effort is spent fighting the codebase instead of building value.</li><li><strong>Bug Bashes Become Bloodbaths:</strong> Production stability plummets. You’re constantly reacting to fires because the foundations are made of gasoline-soaked rags. You discover critical bugs months after deployment, often from angry customer support tickets.</li><li><strong>Onboarding is Hell:</strong> Try explaining that uncommented, untested, thousand-line function to a new hire. Watch their initial enthusiasm curdle into quiet despair.</li><li><strong>Innovation Dies:</strong> Forget cool new features. Your best engineers are perpetually tied up keeping the leaky ship afloat.</li></ul><p><strong>But Wait, There Are “Solutions”! (Spoiler: They were not solutions.)</strong></p><p>When the pain becomes undeniable (but the root cause remains ignored), we enter the magical realm of corporate coping mechanisms:</p><ul><li><strong>Deploy the QA Army:</strong> “If we just hire enough testers to manually find <em>every single problem</em> before release, it’s cheaper than writing stable code, right? Right?!” (Spoiler: No. It’s reactive, expensive, and doesn’t scale.)</li><li><strong>The Offshore Brute-Force Brigade:</strong> “Let’s hire 10x the engineers in a cheaper location to just throw bodies at the problem!” (Check back in 2 years when the codebase is an eldritch horror, costs have spiraled, and you need a full rewrite anyway.)</li><li><strong>Declare Bankruptcy on Reality:</strong> “The debt doesn’t exist! The slowdown is… uh… market forces! Yes! Our developers are just… less motivated!” (Meanwhile, development costs quietly inflate by 50%, and predictability vanishes.)</li></ul><blockquote><strong>Generating Debt is Normal. Ignoring It is Negligence.</strong></blockquote><p>Look, <em>every</em> company accumulates tech debt. Sometimes it’s a conscious, strategic decision (“Let’s ship this MVP fast, knowing we’ll refactor in Q3”). Sometimes architecture naturally ages as the business evolves. That’s fine! That’s manageable.</p><p>The issues arise when the debt is consistently ignored, dismissed, or treated as an imaginary boogeyman cooked up by whiny engineers. Proactively managing tech debt — identifying it, prioritizing it, and paying it down strategically — is <em>essential</em> for maintaining stable, predictable development and, ultimately, a healthy business.</p><p>So, to the C-level executive who doesn’t “give a sh*t” about tech debt: You might not care <em>now</em>. But the debt is real. It’s accruing interest. And telling the engineers wrestling with its consequences every single day that their reality is bullshit? That’s not just ignorant. It’s profoundly insulting.</p><p>The bill always comes due. Always. And the collectors (in the form of system outages, missed deadlines, spiraling costs, and burnt-out engineers) are very, very patient. Good luck.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e300d009658a" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>