<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Julio Daniel Reyes</title>
    <description>The latest articles on DEV Community by Julio Daniel Reyes (@nemesiscodex).</description>
    <link>https://dev.to/nemesiscodex</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F291938%2Fa18aad0a-8548-4b0f-b389-a81e144546af.png</url>
      <title>DEV Community: Julio Daniel Reyes</title>
      <link>https://dev.to/nemesiscodex</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/nemesiscodex"/>
    <language>en</language>
    <item>
      <title>How to Make Your AI Coding Agent Prove Its Work</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Thu, 29 Oct 2026 03:16:47 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/how-to-make-your-ai-coding-agent-prove-its-work-1617</link>
      <guid>https://dev.to/nemesiscodex/how-to-make-your-ai-coding-agent-prove-its-work-1617</guid>
      <description>&lt;p&gt;Your AI agent says "Done." The build is green. The tests pass.&lt;/p&gt;

&lt;p&gt;Did it actually work? Did it do the correct thing?&lt;/p&gt;

&lt;p&gt;Most of the time, you only find out when you check it yourself. That makes you the verifier for every task, and the agent is only as fast as your review.&lt;/p&gt;

&lt;p&gt;I spent the last few weeks with &lt;strong&gt;pstack&lt;/strong&gt;, a plugin by Lauren Tan (poteto) that changes this. I turned the official guide into a 19-minute video and 24 Shorts. In this post, I will teach you the core habit, show you the prompts that make it work, and point you to the part of the video for each topic.&lt;/p&gt;

&lt;h2&gt;
  
  
  TL;DR
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;If you remember one thing: give the agent a goal and a way to check it.&lt;/strong&gt;&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;/poteto-mode the export writes duplicate rows when a retry lands mid-run. repro first, then fix and verify.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Install pstack&lt;/strong&gt; and run &lt;code&gt;/setup-pstack&lt;/code&gt;. Say yes when it offers a verification skill.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Write prompts that can fail.&lt;/strong&gt; A goal, a done check, and the proof you want to see.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Steer with principle names.&lt;/strong&gt; "apply prove it works" redirects the agent better than a paragraph.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  The Problem
&lt;/h2&gt;

&lt;p&gt;Here is a prompt most of us have typed:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;fix the export bug
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Can you spot what's wrong?&lt;/p&gt;

&lt;p&gt;There is no way to know when it is done. The agent will change some code, run the build, and report success. A green build proves the code compiles. It does not prove the export stopped writing duplicate rows.&lt;/p&gt;

&lt;p&gt;So you open the app, run the export, and count rows yourself. Every task ends with you checking by hand.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The scary part is that this gets worse as agents get faster.&lt;/strong&gt; More output with no proof only means more unchecked work for you.&lt;/p&gt;

&lt;h2&gt;
  
  
  The Solution
&lt;/h2&gt;

&lt;p&gt;pstack is a set of skills, playbooks, and principles for AI coding agents. It started in Cursor, and it now installs in Claude Code too.&lt;/p&gt;

&lt;p&gt;The entry point is one command, &lt;code&gt;/poteto-mode&lt;/code&gt;. You give it a goal. It matches one of 23 playbooks, copies that playbook's steps into a todo list, and calls the other skills as each step needs them:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;your prompt
└── /poteto-mode
    ├── question           → Investigation ─┐
    ├── defect             → Bug fix        ├─→ verify and show evidence
    ├── new behavior       → Feature        │
    ├── structure only     → Refactoring    │
    └── measured slowness  → Perf issue    ─┘
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Notice how every route ends in the same place. The reply carries the evidence: the commands run, their output, the stored value, the row count. You read proof instead of producing it.&lt;/p&gt;

&lt;p&gt;Now the bad prompt becomes this:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;/poteto-mode the csv export drops its last row since yesterday's deploy. failing job id is 4812. repro first, then fix. done means the 60k-row fixture exports every row. show me the row counts before and after.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;It is longer, but every part does work. And once the conversation has context, your follow-ups shrink to &lt;code&gt;continue&lt;/code&gt; or &lt;code&gt;keep going until done&lt;/code&gt;.&lt;/p&gt;

&lt;p&gt;Here is the full walkthrough, from setup to overnight runs:&lt;/p&gt;

&lt;p&gt;  &lt;iframe src="https://www.youtube.com/embed/UXcMa3fOHXk" width="710" height="399"&gt;
  &lt;/iframe&gt;
&lt;/p&gt;

&lt;h2&gt;
  
  
  How to Use It
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1) Install and pick your models
&lt;/h3&gt;

&lt;p&gt;In Cursor:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;/add-plugin pstack
/setup-pstack
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;In Claude Code, add the marketplace, install, and then every pstack command takes a &lt;code&gt;pstack:&lt;/code&gt; prefix:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;/plugin marketplace add cursor/plugins
/plugin install pstack@cursor-plugins
/pstack:setup-pstack
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;&lt;code&gt;/setup-pstack&lt;/code&gt; detects your models, asks for a reasoning budget, and writes a small rule file every skill reads. At the end it offers to create a &lt;strong&gt;verification skill&lt;/strong&gt; for your project. Say yes.&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;&lt;strong&gt;Important:&lt;/strong&gt; The verification skill is the part that pays off the most. An agent that can drive your app checks its own work and keeps going until the check passes. An agent that can't hands every result back to you.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;In the video: &lt;a href="https://www.youtube.com/watch?v=UXcMa3fOHXk&amp;amp;t=26s" rel="noopener noreferrer"&gt;setup at 0:26&lt;/a&gt;, &lt;a href="https://www.youtube.com/watch?v=UXcMa3fOHXk&amp;amp;t=633s" rel="noopener noreferrer"&gt;the verification skill at 10:33&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  2) Write a prompt that can pass or fail
&lt;/h3&gt;

&lt;p&gt;A useful prompt carries up to five things:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;The goal.&lt;/strong&gt; What is wrong, or what you want.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The done check.&lt;/strong&gt; It must be able to fail. "Make it better" is not a check.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The proof you want.&lt;/strong&gt; Command output, a video of the flow, a before and after number.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;What you already know.&lt;/strong&gt; A log line, a job id, a repro step.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The real constraints.&lt;/strong&gt; "repro first", "zero behavior change", "don't change any code yet".&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Leave out two things: &lt;strong&gt;the how&lt;/strong&gt;, and &lt;strong&gt;your theory of the cause&lt;/strong&gt;. A stated guess narrows the search to wherever you pointed. Let the agent restate the problem first:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;/poteto-mode read this thread. restate the underlying issue in your own words, in plain english. don't change any code yet.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;A misreading shows up in the restatement, before any code exists. In the video: &lt;a href="https://www.youtube.com/watch?v=UXcMa3fOHXk&amp;amp;t=159s" rel="noopener noreferrer"&gt;writing a good prompt at 2:39&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  3) Steer with principle names
&lt;/h3&gt;

&lt;p&gt;pstack ships 24 principles. You never call them directly. &lt;code&gt;/poteto-mode&lt;/code&gt; reads them on every multi-step task, and you use their names to redirect the agent mid-task:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;apply prove it works. run the real import flow and show me the written records.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;





&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;use subtract before you add. delete the obsolete adapters first, then design what's left.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;One name points at a full rule the agent already read. That is why a single line is enough.&lt;/p&gt;

&lt;p&gt;This is what the 24 Shorts are for. Each one is about 80 seconds: the rule, a small example, and how to tell if the agent applied it. Good ones to start with:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;a href="https://www.youtube.com/shorts/U7VzxLb14aw" rel="noopener noreferrer"&gt;#17 Prove It Works&lt;/a&gt;: check the real thing, not a proxy, a self report, or "it compiles".&lt;/li&gt;
&lt;li&gt;
&lt;a href="https://www.youtube.com/shorts/ArdafTwYdQc" rel="noopener noreferrer"&gt;#18 Fix Root Causes&lt;/a&gt;: trace the problem to its cause and fix it there.&lt;/li&gt;
&lt;li&gt;
&lt;a href="https://www.youtube.com/shorts/CmnXbuJ3Fkg" rel="noopener noreferrer"&gt;#1 Laziness Protocol&lt;/a&gt;: the most result with the least code and complexity.&lt;/li&gt;
&lt;li&gt;
&lt;a href="https://www.youtube.com/shorts/0WZIlDxMzYY" rel="noopener noreferrer"&gt;#22 Guard the Context Window&lt;/a&gt;: every token in the session should be worth its cost.&lt;/li&gt;
&lt;li&gt;
&lt;a href="https://www.youtube.com/shorts/iFv9_T7PLys" rel="noopener noreferrer"&gt;#24 Encode Lessons in Structure&lt;/a&gt;: turn a repeated correction into a tool, a check, or a lint.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;All 24 are in &lt;a href="https://www.youtube.com/playlist?list=PLPtrbYFRMIqk" rel="noopener noreferrer"&gt;one playlist&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  4) Leave it running overnight
&lt;/h3&gt;

&lt;p&gt;This is the payoff. An agent that verifies its own work is an agent you can leave alone with a hard task. The handoff does not need to be long:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;/poteto-mode im going to bed. migrate every caller to the new parser in a fresh worktree off main.
done means zero old callers, all parser fixtures pass, old api deleted.
keep a decision log. don't ask me before committing.
/loop until done. if you're truly stuck after a few hours, stop and write up why.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Each line buys you something:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;"im going to bed"&lt;/strong&gt; tells the agent to stop asking and keep going.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;"done means..."&lt;/strong&gt; gives every loop iteration a check it can run.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;"fresh worktree"&lt;/strong&gt; keeps it away from anything else you have open.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;"keep a decision log"&lt;/strong&gt; gives you something to audit in the morning.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;The escape hatch&lt;/strong&gt; lets it stop at a real dead end instead of reinterpreting the goal for eight hours.&lt;/li&gt;
&lt;/ol&gt;

&lt;blockquote&gt;
&lt;p&gt;&lt;strong&gt;Caveat:&lt;/strong&gt; A loop you don't trust only produces unchecked work faster. Do the task once while you watch, and get the verification skill working, before you walk away.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;In the video: &lt;a href="https://www.youtube.com/watch?v=UXcMa3fOHXk&amp;amp;t=711s" rel="noopener noreferrer"&gt;the overnight chapters start at 11:51&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  Tradeoffs
&lt;/h2&gt;

&lt;p&gt;pstack is not free. Here is what you pay.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Pros&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Replies carry evidence, so you review proof instead of re-running everything.&lt;/li&gt;
&lt;li&gt;Short prompts work, because the playbook holds the structure.&lt;/li&gt;
&lt;li&gt;Principle names give you a precise way to correct a drifting agent.&lt;/li&gt;
&lt;li&gt;Safe long runs become possible once verification works.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;Cons&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;More tokens.&lt;/strong&gt; Subagents and review panels cost money. &lt;code&gt;/setup-pstack&lt;/code&gt; lets you pick a smaller budget or cheaper models.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Slower on small edits.&lt;/strong&gt; A one-line change does not need a playbook. Save &lt;code&gt;/poteto-mode&lt;/code&gt; for work that needs rigor.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Setup work.&lt;/strong&gt; The verification skill needs a way to drive your app, and that takes a first run to get right.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Best Practices
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Don't list skills in the prompt.&lt;/strong&gt; "use /how then /architect then /arena" fights the order the playbook already has. State the goal; name a skill only to override a default.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Never accept a green build as proof.&lt;/strong&gt; Ask for the real command, flow, or stored value.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Ask for prototypes before you pick a design.&lt;/strong&gt; The first design is only the first shape the model thought of.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;One worktree per parallel agent.&lt;/strong&gt; Otherwise they overwrite each other.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Vet numbers before you share them.&lt;/strong&gt; A warm cache can fake a speedup; run &lt;code&gt;/benchmark-checklist&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Fix the repo, not the chat.&lt;/strong&gt; &lt;code&gt;/correct&lt;/code&gt; turns a repeated mistake into a check, so the next run can't repeat it.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;The video ends with &lt;a href="https://www.youtube.com/watch?v=UXcMa3fOHXk&amp;amp;t=1090s" rel="noopener noreferrer"&gt;6 common mistakes at 18:10&lt;/a&gt;, if you want them in under a minute.&lt;/p&gt;

&lt;h2&gt;
  
  
  Conclusion
&lt;/h2&gt;

&lt;p&gt;pstack does not make the model smarter. It changes what "done" means. &lt;strong&gt;A goal plus a way to check it&lt;/strong&gt; is the whole idea, and everything else in the plugin builds on that.&lt;/p&gt;

&lt;p&gt;Start with one real, small task:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;/poteto-mode add a --json flag to this command. text output stays byte-identical. verify both.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Then watch the &lt;a href="https://www.youtube.com/watch?v=UXcMa3fOHXk" rel="noopener noreferrer"&gt;full guide&lt;/a&gt; for the rest: understanding code with &lt;code&gt;/how&lt;/code&gt; and &lt;code&gt;/why&lt;/code&gt;, comparing designs with &lt;code&gt;/arena&lt;/code&gt;, landing a stack of PRs, and building your own mode.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;pstack was created by Lauren Tan (poteto). My videos are an unofficial explainer based on the &lt;a href="https://github.com/cursor/plugins/tree/main/pstack/docs/guide" rel="noopener noreferrer"&gt;official guide&lt;/a&gt;; they are not made by or affiliated with Cursor. I drafted this post with help from an AI agent and reviewed it myself.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;If you made it this far, thank you for reading. What evidence do you ask your coding agent for before you trust a change?&lt;/p&gt;

</description>
      <category>ai</category>
      <category>programming</category>
      <category>productivity</category>
      <category>cursor</category>
    </item>
    <item>
      <title>I made a NES-style 8-bit sequencer app</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Wed, 01 Apr 2026 18:16:11 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/i-made-a-nes-style-8-bit-sequencer-app-2g38</link>
      <guid>https://dev.to/nemesiscodex/i-made-a-nes-style-8-bit-sequencer-app-2g38</guid>
      <description>&lt;p&gt;Just for fun!&lt;/p&gt;

&lt;p&gt;Ever wondered what 8-bit music actually is?&lt;/p&gt;

&lt;p&gt;I started asking myself: why 8-bit? What gives it that distinctive sound?&lt;/p&gt;

&lt;p&gt;I learned that 8-bit music is mostly a nickname for the sound associated with old game hardware. Those systems used 8-bit processors, had very limited memory, and could not store and play full recorded songs the way modern systems do. Instead, they used dedicated audio hardware (an APU) to generate sound in real time through a small number of voices (channels).&lt;/p&gt;

&lt;p&gt;Each voice could only play one note at a time.&lt;/p&gt;

&lt;p&gt;The NES APU had 5 channels:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;2 pulse voice (square-like waves): melody / harmony&lt;/li&gt;
&lt;li&gt;1 triangle voice: bass&lt;/li&gt;
&lt;li&gt;1 noise voice: percussion / effects&lt;/li&gt;
&lt;li&gt;1 DPCM voice: short samples&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;That sounded simple and fun enough that I wanted to try building something inspired by it.&lt;/p&gt;

&lt;p&gt;So I built my own browser-based NES-style sequencer: &lt;a href="https://ochoit.nemesiscodex.org" rel="noopener noreferrer"&gt;Ocho(8)-it!&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Open source and MIT licensed.&lt;/p&gt;

</description>
      <category>music</category>
      <category>webapp</category>
      <category>gamedev</category>
      <category>opensource</category>
    </item>
    <item>
      <title>Stop Sending "Monster" PRs: Use stacked pull requests</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Mon, 26 Jan 2026 18:49:06 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/stop-sending-monster-prs-use-stacked-pull-requests-4kf8</link>
      <guid>https://dev.to/nemesiscodex/stop-sending-monster-prs-use-stacked-pull-requests-4kf8</guid>
      <description>&lt;p&gt;Ever worked on a &lt;em&gt;"quick"&lt;/em&gt; feature only to realize it involves changes across your data model, services, APIs, and frontend?&lt;/p&gt;

&lt;p&gt;You create a single branch, dump 30 commits into it, and open a PR so big your team needs a weekend retreat to review it.&lt;/p&gt;

&lt;p&gt;Or even worse, did you ever have to review one of those monster PRs?&lt;/p&gt;

&lt;p&gt;That's how you end up with developer fatigue.&lt;/p&gt;

&lt;p&gt;Reviewing a large PR takes time if you want quality, yet you're blocking your teammates until it's done.&lt;/p&gt;

&lt;p&gt;The trade-off between speed and review depth is real.&lt;/p&gt;

&lt;p&gt;You may think, "It's just one PR, no big deal." But what if that becomes the norm for everyone on the team?&lt;/p&gt;

&lt;h3&gt;
  
  
  So what's the alternative?
&lt;/h3&gt;

&lt;p&gt;&lt;strong&gt;Stacked Pull Requests&lt;/strong&gt;: break the work into tiny, reviewable chunks that land in sequence.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fmad6h01ecis8gaqukn3g.png" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fmad6h01ecis8gaqukn3g.png" alt="Every chunk stacked on top of it's dependency" width="800" height="661"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;In this video I show the workflow, the manual pain points, and how tools like the Graphite CLI automate the stacking so you can ship faster without the maintenance headaches.&lt;/p&gt;

&lt;h3&gt;
  
  
  Links:
&lt;/h3&gt;

&lt;ul&gt;
&lt;li&gt;Graphite CLI: &lt;a href="https://graphite.com/docs/cli-overview" rel="noopener noreferrer"&gt;https://graphite.com/docs/cli-overview&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;How to structure your stacks: &lt;a href="https://graphite.com/docs/how-to-structure-your-stacks" rel="noopener noreferrer"&gt;https://graphite.com/docs/how-to-structure-your-stacks&lt;/a&gt;
&lt;/li&gt;
&lt;li&gt;Best practices for reviewing Stacked PRs: &lt;a href="https://graphite.com/docs/best-practices-for-reviewing-stacks" rel="noopener noreferrer"&gt;https://graphite.com/docs/best-practices-for-reviewing-stacks&lt;/a&gt;
&lt;/li&gt;
&lt;/ul&gt;

</description>
      <category>programming</category>
      <category>git</category>
      <category>productivity</category>
      <category>learning</category>
    </item>
    <item>
      <title>A Better Way to Run Git Worktrees Finally!</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Tue, 16 Dec 2025 13:35:04 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/a-better-way-to-run-git-worktrees-finally-1lh9</link>
      <guid>https://dev.to/nemesiscodex/a-better-way-to-run-git-worktrees-finally-1lh9</guid>
      <description>&lt;p&gt;I recently found git-gtr, a small CLI by the CodeRabbit team that wraps native Git worktrees and makes them practical for daily use.&lt;/p&gt;

&lt;p&gt;Repository: &lt;a href="https://github.com/coderabbitai/git-worktree-runner" rel="noopener noreferrer"&gt;https://github.com/coderabbitai/git-worktree-runner&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Install it, create an alias for it &lt;code&gt;alias gw="git gtr"&lt;/code&gt; (Easier to remember, but up to you!).&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Navigate to your repository&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;gw new &amp;lt;branchName&amp;gt;&lt;/code&gt; will create a new worktree. The first time it will create a new folder in the parent directory &lt;code&gt;&amp;lt;repo&amp;gt;-worktrees&lt;/code&gt;, inside it a folder &lt;code&gt;&amp;lt;branchName&amp;gt;&lt;/code&gt; for your newly created worktree.&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;gw editor &amp;lt;branchName&amp;gt;&lt;/code&gt; will open the configured code editor.&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;gw ai &amp;lt;branchName&amp;gt;&lt;/code&gt; will start your cli (claude, codex, opencode)&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;gw run &amp;lt;branchName&amp;gt; &amp;lt;command&amp;gt;&lt;/code&gt; will run commands on that worktree&lt;/li&gt;
&lt;li&gt;
&lt;code&gt;gw rm &amp;lt;branchName&amp;gt;&lt;/code&gt; to remove it once you are done.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Full demo in my new vid. Would you use this? or do you prefer the original git command?&lt;/p&gt;

</description>
      <category>git</category>
      <category>vibecoding</category>
      <category>programming</category>
      <category>ai</category>
    </item>
    <item>
      <title>(Video) Authentication Service in Actix - Part 1: Configuration</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Mon, 06 Jul 2020 05:13:58 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/video-authentication-service-in-actix-part-1-configuration-45a9</link>
      <guid>https://dev.to/nemesiscodex/video-authentication-service-in-actix-part-1-configuration-45a9</guid>
      <description>&lt;p&gt;&lt;iframe width="710" height="399" src="https://www.youtube.com/embed/AH2P7Vc0N9s"&gt;
&lt;/iframe&gt;
&lt;/p&gt;

&lt;p&gt;In this video series, I'm going to build an authentication service in Actix.&lt;/p&gt;

&lt;p&gt;There are a couple of new libraries that I want to try, and a few people already asked me to do authentication, so let's see how it goes.&lt;/p&gt;

&lt;p&gt;The idea is to have a service that will allow the user to create an account, log in and have access to other secure resources. I'm going to do something simple using only JSON Web Tokens, but it will give you an idea of how to implement a more complex setup like combining it with Oauth 2. &lt;/p&gt;

&lt;p&gt;In this first part, I will set up the app configuration, the logs, and add a simple health endpoint.&lt;/p&gt;

</description>
      <category>rust</category>
      <category>webdev</category>
      <category>tutorial</category>
      <category>security</category>
    </item>
    <item>
      <title>Rust Web development | Boilerplate free with Rocket</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Wed, 03 Jun 2020 07:14:23 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/rust-web-development-boilerplate-free-with-rocket-438l</link>
      <guid>https://dev.to/nemesiscodex/rust-web-development-boilerplate-free-with-rocket-438l</guid>
      <description>&lt;p&gt;&lt;iframe width="710" height="399" src="https://www.youtube.com/embed/tjH0Mye8U_A"&gt;
&lt;/iframe&gt;
&lt;/p&gt;

&lt;p&gt;An overview of the Rocket web framework for rust.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Type Safe&lt;/li&gt;
&lt;li&gt;Boilerplate Free&lt;/li&gt;
&lt;li&gt;Easy to use&lt;/li&gt;
&lt;li&gt;Extensible&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Including examples:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Create a Hello World service &lt;/li&gt;
&lt;li&gt;Create a TODO service with a Postgres database using Diesel ORM.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;⏱️TIMESTAMPS⏱️&lt;br&gt;
0:00 Rust Web development with Rocket&lt;br&gt;
0:07 What is Rocket?&lt;br&gt;
0:49 Rocket overview&lt;br&gt;
3:32 A couple of things to consider before using Rocket&lt;br&gt;
4:22 Building a Hello World service&lt;br&gt;
8:37 Define custom configuration&lt;br&gt;
9:29 Building a TODO service with a Postgres database&lt;br&gt;
9:34 Setup a Database service using Docker and Docker Compose&lt;br&gt;
10:12 Diesel to manage database migrations&lt;br&gt;
12:19 Diesel and Serde dependencies&lt;br&gt;
12:58 Rearrange the endpoints&lt;br&gt;
13:20 Use the Schema file generated by Diesel&lt;br&gt;
13:47 Define the models for the TODO service&lt;br&gt;
15:04 Define handler to create new TODOs&lt;br&gt;
15:52 How to get a DB connection with Rocket?&lt;br&gt;
17:09 Create new TODO&lt;br&gt;
18:00 Define handler to list all TODOs&lt;br&gt;
18:54 Define the database configuration&lt;br&gt;
19:20 Testing the TODO endpoints&lt;br&gt;
20:27 Bonus: Define endpoint to mark a TODO as checked&lt;/p&gt;

&lt;p&gt;📔 Source Code&lt;br&gt;
&lt;a href="https://github.com/nemesiscodex/hello-rocket"&gt;https://github.com/nemesiscodex/hello-rocket&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;🚀 Rocket Web Framework&lt;br&gt;
&lt;a href="https://rocket.rs/"&gt;https://rocket.rs/&lt;/a&gt;&lt;/p&gt;

</description>
      <category>rust</category>
      <category>webdev</category>
      <category>tutorial</category>
      <category>postgres</category>
    </item>
    <item>
      <title>#3 Building a simple GraphQL API with Actix and Juniper (Rust)</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Mon, 11 May 2020 00:48:03 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/3-building-a-simple-graphql-api-with-actix-and-juniper-rust-31pm</link>
      <guid>https://dev.to/nemesiscodex/3-building-a-simple-graphql-api-with-actix-and-juniper-rust-31pm</guid>
      <description>&lt;p&gt;&lt;iframe width="710" height="399" src="https://www.youtube.com/embed/7v7ERnrC4fo"&gt;
&lt;/iframe&gt;
&lt;/p&gt;

&lt;p&gt;⏱️TIMESTAMPS⏱️&lt;br&gt;
0:00 Simple GraphQL API with Actix and Juniper - part 3&lt;br&gt;
0:07 Start of the video&lt;br&gt;
0:28 Refactoring of User queries into a repository&lt;br&gt;
4:06 Get a single user&lt;br&gt;
6:45 Code to create and list Posts&lt;br&gt;
11:15 Query the posts on GraphiQL&lt;br&gt;
13:10 Posts as a field of User&lt;br&gt;
15:59 N+1 Problem&lt;br&gt;
16:45 Solution? Dataloaders&lt;br&gt;
17:11 Using a Dataloader to get the posts of an user&lt;br&gt;
26:05 Did it work?&lt;br&gt;
27:39 Thank you!&lt;/p&gt;

&lt;p&gt;&lt;a href="https://dev.to/nemesiscodex/building-a-simple-graphql-api-with-actix-and-juniper-rust-26f6"&gt;Part 1&lt;/a&gt;&lt;br&gt;
&lt;a href="https://dev.to/nemesiscodex/2-building-a-simple-graphql-api-with-actix-and-juniper-rust-51h9"&gt;Part 2&lt;/a&gt;&lt;/p&gt;

</description>
      <category>rust</category>
      <category>graphql</category>
      <category>tutorial</category>
      <category>webdev</category>
    </item>
    <item>
      <title>My development environment | From scratch | Ubuntu | 2020</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Thu, 16 Apr 2020 01:38:52 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/my-development-environment-from-scratch-ubuntu-2020-41h9</link>
      <guid>https://dev.to/nemesiscodex/my-development-environment-from-scratch-ubuntu-2020-41h9</guid>
      <description>&lt;p&gt;&lt;iframe width="710" height="399" src="https://www.youtube.com/embed/smwpadzkWRQ"&gt;
&lt;/iframe&gt;
&lt;/p&gt;

&lt;p&gt;Content:&lt;br&gt;
0:07 Start of the video&lt;br&gt;
0:51 Terminals&lt;br&gt;
0:51    - Terminator&lt;br&gt;
2:16    - Alacritty&lt;br&gt;
3:12 Fundamental tools installation - with brief explanation&lt;br&gt;
3:19    - curl&lt;br&gt;
3:28    - git&lt;br&gt;
3:33    - htop&lt;br&gt;
3:39    - vim&lt;br&gt;
3:45    - jq&lt;br&gt;
3:50    - zsh&lt;br&gt;
4:03    - tmux&lt;br&gt;
4:27 Tmux basic usage&lt;br&gt;
5:56 jq overview&lt;br&gt;
6:29 htop overview&lt;br&gt;
6:39 vim basic usage&lt;br&gt;
7:24 zsh overview&lt;br&gt;
7:50 oh my zsh installation&lt;br&gt;
8:58 powerlevel 10k installation (zsh theme)&lt;br&gt;
12:42 powerlevel 10k overview&lt;br&gt;
13:34 fzf installation (fuzzy finder)&lt;br&gt;
14:11 fzf basic usage&lt;br&gt;
15:24 IDEs&lt;br&gt;
15:24    - Intellij Idea (mention)&lt;br&gt;
15:42    - VS Code (mention and installation)&lt;br&gt;
15:59 Nord color theme &lt;br&gt;
16:13 Nord for alacritty&lt;br&gt;
16:53 Jetbrain's Mono Font&lt;br&gt;
17:37 Rust installation&lt;br&gt;
18:00 Customize status bar configuration, add crab emoji (very important)&lt;br&gt;
18:38 Docker installation&lt;br&gt;
20:52 docker-compose installation&lt;/p&gt;

</description>
      <category>linux</category>
      <category>ubuntu</category>
      <category>webdev</category>
      <category>tutorial</category>
    </item>
    <item>
      <title>#2 Building a simple GraphQL API with Actix and Juniper (Rust)</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Mon, 06 Apr 2020 23:09:53 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/2-building-a-simple-graphql-api-with-actix-and-juniper-rust-51h9</link>
      <guid>https://dev.to/nemesiscodex/2-building-a-simple-graphql-api-with-actix-and-juniper-rust-51h9</guid>
      <description>&lt;p&gt;&lt;iframe width="710" height="399" src="https://www.youtube.com/embed/aEAz5DHhpLo"&gt;
&lt;/iframe&gt;
&lt;/p&gt;

&lt;p&gt;&lt;a href="https://dev.to/nemesiscodex/building-a-simple-graphql-api-with-actix-and-juniper-rust-26f6"&gt;Part 1&lt;/a&gt;&lt;/p&gt;

</description>
      <category>rust</category>
      <category>graphql</category>
      <category>tutorial</category>
      <category>webdev</category>
    </item>
    <item>
      <title>Building a simple GraphQL API with Actix and Juniper (Rust)</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Fri, 20 Mar 2020 05:16:02 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/building-a-simple-graphql-api-with-actix-and-juniper-rust-26f6</link>
      <guid>https://dev.to/nemesiscodex/building-a-simple-graphql-api-with-actix-and-juniper-rust-26f6</guid>
      <description>&lt;p&gt;&lt;iframe width="710" height="399" src="https://www.youtube.com/embed/9q4GcWbAIEM"&gt;
&lt;/iframe&gt;
 &lt;/p&gt;

</description>
      <category>rust</category>
      <category>graphql</category>
      <category>tutorial</category>
      <category>webdev</category>
    </item>
    <item>
      <title>Running Actix integration tests with Github actions (Rust) </title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Sat, 07 Mar 2020 11:59:19 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/running-actix-integration-tests-with-github-actions-rust-np7</link>
      <guid>https://dev.to/nemesiscodex/running-actix-integration-tests-with-github-actions-rust-np7</guid>
      <description>&lt;p&gt;&lt;iframe width="710" height="399" src="https://www.youtube.com/embed/TKCJl6YgsH8"&gt;
&lt;/iframe&gt;
 &lt;/p&gt;

</description>
      <category>rust</category>
      <category>devops</category>
      <category>webdev</category>
      <category>testing</category>
    </item>
    <item>
      <title>Error handling and Logging with Actix (Rust)</title>
      <dc:creator>Julio Daniel Reyes</dc:creator>
      <pubDate>Mon, 24 Feb 2020 21:05:27 +0000</pubDate>
      <link>https://dev.to/nemesiscodex/error-handling-and-logging-with-actix-rust-3eni</link>
      <guid>https://dev.to/nemesiscodex/error-handling-and-logging-with-actix-rust-3eni</guid>
      <description>&lt;p&gt;&lt;a href="https://youtu.be/NHaWAWLgtAw"&gt;&lt;img src="https://res.cloudinary.com/practicaldev/image/fetch/s--AsdB_mnF--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_880/https://img.youtube.com/vi/NHaWAWLgtAw/maxresdefault.jpg" alt="Error handling and Logging with Actix (Rust)"&gt;&lt;/a&gt;&lt;/p&gt;

</description>
      <category>rust</category>
      <category>webdev</category>
      <category>microservices</category>
      <category>tutorial</category>
    </item>
  </channel>
</rss>
