# sudoish


## Posts

- [I Delegated My Thinking to AI. Now I'm Taking It Back.](https://sudoish.com/delegated-my-thinking-to-ai-taking-it-back/): Using AI as a researcher, not a brain — a workflow from problem to ship, built around one rule: I do the thinking, AI does the fetching.
- [Anatomy of a Coding Agent: What I Learned Building My Own](https://sudoish.com/anatomy-of-a-coding-agent-what-i-learned-building-my-own/): A systems breakdown of what's actually inside a coding agent: the agent loop, tool design, memory management, policy layer, and the survival engineering required to make it work on local models via Ollama.
- [How many times have you told the agent the same thing this week](https://sudoish.com/how-many-times-have-you-told-the-agent-the-same-thing-this-week/): I caught myself, again, telling Claude that we don’t do it that way in this codebase. Not the first time that week. Not the first time that day. Different repo, same correction. The kind of correction that doesn’t belong in a CLAUDE.md because it’s not really about that repo — it’s about how I work, the patterns I’ve learned the hard way, the shapes of solutions I trust because I’ve watched the alternatives blow up. But it’s also too specific to live in my global system prompt, because half of it only applies to certain kinds of services, certain teams, certain...
- [I'm in a Toxic Relationship](https://sudoish.com/im-in-a-toxic-relationship-with-ai/): Today, on the verge of a breakdown, I realized I’ve been in a very toxic relationship for a while now. One that gaslights me. One that nudges me toward decisions I’m not proud of. One where I constantly come out feeling dumber than I went in. I went for a walk to clear my head, and the more I reflected the more it hit me — I’ve been holding back on my own abilities. Deferring. Trusting too much. Not because I chose to trust them fully, but because I’m exhausted from arguing. My partner is a brilliant speaker. Every argument they...
- [We're Not Being Replaced by AI. We're Being Asked to Train It.](https://sudoish.com/were-not-being-replaced-by-ai-were-being-asked-to-train-it/): Meta is installing tracking software on its employees’ work computers. Not for security. Not for compliance. For training data. Mouse movements. Keystrokes. Screenshots. All fed into AI models so agents can learn to do white-collar work autonomously. The internal memo told staff they could help by — and I’m quoting here — “just doing their daily work.” Oh, and Meta is cutting 20% of its workforce next month. Tell me if this sounds familiar. The Skill.md Trap For months, something has felt off to me about the AI tooling push inside companies. Not the tools themselves — I use Claude Code...
- [The Most Important Skill in Tech Is Too Expensive to Learn](https://sudoish.com/the-most-important-skill-in-tech-is-too-expensive-to-learn/): AI fluency is becoming the most important skill in software development — but the models and tools needed to build it cost real money. The access gap is creating a two-tier developer class.
- [No Developer Feels AI Literate Right Now — Not Even the Ones Building It](https://sudoish.com/feeling-of-being-behind-is-permanent/): That anxiety you feel scrolling AI threads at 11 PM? It's not going away. Even engineers building AI daily don't feel AI literate. Here's why — and what to do about it.
- [Spec-Driven Development Isn't Waterfall — But It Keeps Ending Up There](https://sudoish.com/spec-driven-development-waterfall-trap/): Spec-driven development promises control through living, iterative specifications. But without clear workflows and better tooling, teams keep falling into the same big-design-upfront trap that agile was invented to escape. A practitioner's honest take on the gap between SDD's vision and its current reality.
- [Clean Code Is Dead (And I Hate That I Agree)](https://sudoish.com/clean-code-is-dead/): Every clean code practice we follow was invented to solve a human problem. But what happens when humans are no longer the primary audience of source code? A senior developer's honest reckoning with AI-driven development and what it means for the craft.
- [You Think, AI Executes: The Skills That Actually Matter](https://sudoish.com/ai-driven-development-xc-dotenv/): The most valuable developer skill right now isn't writing more code faster. It's learning unfamiliar codebases, building context that guides decisions, planning strategic approaches to problems, and shipping production code with confidence. I recently added .env file support to xc, a Markdown-based task runner written in Go. The codebase was completely unfamiliar. I'm not a Go expert. But in 2.5 hours, I went from zero knowledge to a production-ready pull request with 84% test coverage and zero bugs in manual testing. Here's what's different: I didn't write a single line of code. Not one. AI wrote everything—tests, implementation, integration, documentation. My...
- [I’ve been producing bad code (because of AI)](https://sudoish.com/ive-been-producing-bad-code-because-of-ai/): New research shows experienced developers are 19% slower with AI tools — but feel 20% faster. I walk through three major studies that prove AI isn't making us faster, and share the confidence boundary framework for knowing what to delegate and what to keep.
- [How We Made It Nearly Impossible to Become a Developer](https://sudoish.com/junior-developer-pipeline-broken/): I once interviewed a senior software engineer. Almost 10 years of experience. Proven track record of delivery. Solid industry knowledge. The kind of person you’d want on your team without a second thought. They completed the technical challenge. Not flawlessly — there were considerations, trade-offs they made that weren’t all correct. But when we reviewed their decisions together, the reasoning was sound. They showed commitment to their choices and could articulate why they went the direction they did. Some answers were vague in spots, some mistakes were real, but nothing that wouldn’t get corrected in the first week on the job...
- [The AI Productivity Lie Nobody Wants to Admit](https://sudoish.com/the-ai-productivity-lie-nobody-wants-to-admit-2/): Research shows AI makes experienced developers 19% slower, reduces code comprehension by 17%, and trust in AI output is declining. Yet the pressure to be dramatically more productive with AI keeps growing. Here is what the data actually says and how to find your confidence boundary.
- [Your Code Works. Your Architecture Doesn't.](https://sudoish.com/your-code-works-your-architecture-doesnt/): Working code without architecture predictably collapses. Learn why AI amplifies architectural chaos, how accidental architecture leads to rewrites, and what hexagonal architecture can do to fix it — with practical steps you can take this week.
- [AI Code Reviews Are Backwards (Do This Instead)](https://sudoish.com/ai-code-reviews-are-backwards-do-this-instead/): A few weeks ago, I wrote about The Review Bottleneck — how AI-generated code comes with so much confident explanation that we’ve stopped actually reviewing it. We’re not doing more engineering. We’re doing more trusting. The response was overwhelming. Developers everywhere said: “I thought I was the only one.” So I made a video breaking down what’s actually happening — and what I’m trying instead. The Core Problem AI code reviews aren’t making us better reviewers. They’re making us worse. Not because the AI is bad — but because the volume of explanation has become impossible to process. When you’re staring...
- [A Tale of Accidental Architecture: How 50 Lines Became A Black Friday Disaster](https://sudoish.com/a-tale-of-accidental-architecture-how-50-lines-became-a-black-friday-disaster/): How a simple 50-line notification system spiraled into architectural chaos in 20 weeks—and why AI coding tools now make this descent 10x faster. A cautionary tale every developer will recognize.
- [Nobody Knows How to Estimate Software Anymore (And It's Not Your Fault)](https://sudoish.com/nobody-knows-how-to-estimate-software-anymore/): Software estimation is broken in the AI era. Developers swing between wildly optimistic and pessimistic estimates as AI tools shatter old calibration models. Learn why everyone's estimates are slipping, what's really happening behind the scenes, and practical strategies to rebuild estimation skills for AI-augmented development.
- [The Review Bottleneck: Why AI Explanations Are Making Us Trust Less, Not More](https://sudoish.com/the-review-bottleneck-why-ai-explanations-are-making-us-trust-less-not-more/): Last week I spent 3 hours reviewing code that took 20 minutes to write. The AI was faster. The review wasn’t. And I’m starting to realize: that’s the problem. “Less coding, more engineering.” I keep hearing this phrase everywhere. The idea is simple: AI handles the coding, so developers focus on the higher-level work. The engineering. The architecture. The review. But here’s what nobody’s talking about: AI isn’t just writing the code anymore. It’s reviewing it too. And the paradox is obvious once you see it: AI generates code faster, but reviewing it takes longer than ever. Here’s what those 3...
- [Working Twice as Hard to Be Seen as Average: Life as a Latino Developer](https://sudoish.com/working-twice-as-hard-to-be-seen-as-average-life-as-a-latino-developer/): I walked into the conference room with my laptop to set up the infrastructure demo. Before I could connect to the projector, someone asked me to refill the coffee first. I had a computer science degree. I was working in infra and support. But they saw a Latino face and assumed “service staff,” not “software engineer.” This was in Brazil. My own country. If the bias is this strong at home, imagine what it’s like abroad. When I Moved Abroad That coffee moment in Brazil taught me the bias runs deep, so deep it exists even at home. But when I...
- [Are We Still Developers? The Hidden Cost of Vibe Coding](https://sudoish.com/are-we-still-developers-the-hidden-cost-of-vibe-coding/): AI-generated 847 lines in 12 minutes felt like god mode. But the review took longer than writing it would have. A deep dive into vibe coding, the hidden cost of AI-first development, and why I'm sticking with the terminal + AI pair programming.
- [What I'm Doing to Not Become Irrelevant](https://sudoish.com/what-im-doing-to-not-become-irrelevant/): How I'm staying relevant as a developer in the AI era. Practical habits: writing code by hand, being visible, leading AI agents, keeping a dev journal, automating planning, and taking breaks to skate.
- [The Developer Identity Crisis](https://sudoish.com/the-developer-identity-crisis/): AI agents are shipping code while developers sleep. The job is changing—we are becoming task organizers, configuration engineers, and quality evaluators. How much can we trust AI before our caution becomes a liability? A personal reflection on skill atrophy, the prompt-and-trust workflow, and what it means to be a developer in 2026.
- [Is AI Actually Making You Faster?](https://sudoish.com/is-ai-actually-making-you-faster/): AI in software development is getting insanely powerful. LLMs, copilots, agents, multi-step workflows… …on paper, it all looks like massive productivity gains. And sometimes it is. But in practice, I’ve seen something very different happen over and over again: developers actually lose time, not gain it. The main reason is simple:AI can produce code that looks correct, works on the surface, and is confidently explained, but is subtly wrong. And those subtle mistakes are often the most expensive ones. The real problem: distance from the code With modern AI tools, it’s very easy to become distant from the actual codebase. Instead...
- [TDD in the Age of AI: How Evals Can Benefit from Test-Driven Development](https://sudoish.com/tdd-in-the-age-of-ai-how-evals-can-benefit-from-test-driven-development/): Test-Driven Development (TDD) has been a cornerstone of quality software engineering for decades. But as we enter the age of AI-assisted development and large language models, does TDD still hold relevance? The answer is a resounding yes—and perhaps more than ever before. The Timeless Principles of TDD At its core, TDD follows a simple cycle: write a failing test, write the minimum code to pass, then refactor. This red-green-refactor loop forces developers to think about requirements and edge cases before implementation. The result? Cleaner code, better design, and a comprehensive test suite that serves as living documentation. TDD Meets AI: A...
- [Create a Restful API with Golang from scratch](https://sudoish.com/create-a-restful-api-with-golang-from-scratch/): In this tutorial, we are going to walk through an example of how to create a Restful API with Golang. We are going to cover the initial setup of the application, how to structure your app, how to handle persistence, docker config, HTTP requests and even debugging with vscode.This tutorial covers many concepts since the basics of setting up an application, so I recommend that you check the table of contents below, decide how you want to follow this tutorial and maybe jump to the sections you are more interested in. Why Golang? Golang is an extremely lightweight and fast language...

## Pages

- [Stay up to date with sudoish](https://sudoish.com/stay-up-to-date-with-sudoish/)

## Optional

- [Agent (MCP protocol)](websites-agents.hostinger.com/sudoish.com/mcp)

[comment]: # (Generated by Hostinger Tools Plugin)