<?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 La Min Ko on Medium]]></title>
        <description><![CDATA[Stories by La Min Ko on Medium]]></description>
        <link>https://medium.com/@laminko?source=rss-7f84c1b875a------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*K2kSO0aQnEUEDUk4cXBAeQ.jpeg</url>
            <title>Stories by La Min Ko on Medium</title>
            <link>https://medium.com/@laminko?source=rss-7f84c1b875a------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 06:31:49 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@laminko/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[I built SpendCheck to test “vibe coding” with Claude Code]]></title>
            <link>https://medium.com/@laminko/i-built-spendcheck-to-test-vibe-coding-with-claude-code-99a745d32692?source=rss-7f84c1b875a------2</link>
            <guid isPermaLink="false">https://medium.com/p/99a745d32692</guid>
            <category><![CDATA[product-development]]></category>
            <category><![CDATA[claude-code]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[vibe-coding]]></category>
            <dc:creator><![CDATA[La Min Ko]]></dc:creator>
            <pubDate>Wed, 18 Jun 2025 10:30:25 GMT</pubDate>
            <atom:updated>2025-06-18T10:31:13.410Z</atom:updated>
            <content:encoded><![CDATA[<p><em>What started as office banter became a real product in 6 days</em></p><p>I wanted to try Anthropic’s Claude Code and this thing now trending called “vibe coding” — basically building by describing what you want instead of coding from scratch.</p><p>One day at lunch time, my senior mentioned three criteria for fun and hobby app: “Make something that everyone will use, requires minimal attention, and has potential for ads revenue.”</p><p>I borrowed those criteria and brainstormed with Claude. The spending tracker idea emerged — just tap when you spend money, see patterns. I focused on the first two criteria and intentionally forgot about revenue.</p><h3>The experiment</h3><p>Instead of planning and wireframing, I went straight to Claude Code:</p><p>“i want to create simple app, called Spend Check — Just tap if you spent money today, see spending patterns. for frontend, use vue3. for db, use supabase.”</p><p>Complete Vue 3 + Supabase foundation with activity visualization.</p><p>“i want to track the amount too. also show me how can i create a table in supabase”</p><p>Full database schema and amount tracking.</p><p>“ok, i want to use ionic framework with vue, what are the plan, things to be done?”</p><p>Complete migration strategy to mobile-first Ionic Vue app.</p><h3>What surprised me</h3><p>Looking at my 6-day evolution, I never broke flow state. The product improved naturally through conversation:</p><p><strong>Day 1:</strong> “i want to create simple app, called Spend Check — Just tap if you spent money today, see spending patterns”<br><strong>Day 2:</strong> “i want to track the amount too. also show me how can i create a table in supabase”<br><strong>Day 3:</strong> “ok, i want to use ionic framework with vue, what are the plan, things to be done?”<br><strong>Day 4:</strong> “add currency changable” and “add support for Thai baht”<br><strong>Day 5:</strong> “i cannot edit existing entry” → full edit functionality<br><strong>Day 6:</strong> Mobile optimizations and deployment</p><p>Usually I’m constantly googling, debugging configs, reading docs. With vibe coding, I stayed focused on <em>what</em> I wanted, not <em>how</em> to build it.</p><p>My development workflow evolved organically too. Started committing to main, but naturally began creating GitHub issues, branching out, implementing with Claude’s help, and merging through pull requests.</p><p>The conversations were pure natural language: “i cannot edit existing entry”, “add currency changable”, “add support for Thai baht”. Claude handled all technical implementation while I focused on user experience.</p><h3>The result</h3><p><a href="https://spend-check-eta.vercel.app/">SpendCheck</a> went from office joke to deployed app in 6 days. The code quality was better than my usual manual work — Claude knew accessibility patterns and optimizations I would have missed.</p><h3>What I learned</h3><p>Vibe coding isn’t about avoiding code — it’s about staying in creative flow. When you describe what you want and get working implementations instantly, you spend time on what actually matters: how it feels to use.</p><p>What surprised me most was how Claude kept me in the loop whenever there was ambiguity. Thanks to <a href="https://docs.anthropic.com/en/docs/claude-code/memory">Claude’s Memory</a>, I had set a rule that it should ask clarifying questions instead of making assumptions, and it worked perfectly. Instead of guessing, it would ask: “Should users be able to select specific time of day, or just the date?” or “How should we handle future dates — disable them or show an error?”</p><p>This human-in-the-loop approach meant I was making design decisions while Claude handled implementation. Perfect collaboration.</p><p>I think we’re hitting an inflection point where developers can focus more on product thinking. Instead of getting lost in implementation details, vibe coding lets us spend time on product design and management decisions — understanding user needs, making UX choices, and thinking about business value while AI handles the technical execution.</p><p><strong>SpendCheck</strong>: <a href="https://spend-check-eta.vercel.app/">spend-check-eta.vercel.app</a><br><strong>Code</strong>: <a href="https://github.com/laminko/SpendCheck">github.com/laminko/SpendCheck</a></p><p>#VibeCoding #ClaudeCode #AI #ProductDevelopment</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=99a745d32692" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Getting last arguments from previous command: $_ and !$]]></title>
            <link>https://medium.com/@laminko/getting-last-arguments-from-previous-command-and-b53650e6e683?source=rss-7f84c1b875a------2</link>
            <guid isPermaLink="false">https://medium.com/p/b53650e6e683</guid>
            <category><![CDATA[bash]]></category>
            <category><![CDATA[linux]]></category>
            <category><![CDATA[shell]]></category>
            <category><![CDATA[macos]]></category>
            <dc:creator><![CDATA[La Min Ko]]></dc:creator>
            <pubDate>Wed, 08 Jan 2025 18:48:48 GMT</pubDate>
            <atom:updated>2025-01-08T18:55:56.324Z</atom:updated>
            <content:encoded><![CDATA[<h3>Getting last arguments from previous command: $_ and !$</h3><p>To recall and reuse last argument from previous command in the current shell command, you can use these: $_ and !$. I am not sure, but there maybe other ways to do it but I am fine with them.</p><p>Honestly, I knew this function because I saw a glance while my senior was using this a lot in his linux commands for manipulating various *nix command in a magical way. I also wanted to learn and use it.</p><p>I googled it and found out:</p><ul><li>$_ is a shell variable, meaning its value is available within the current shell environment. It stores the last argument.</li><li>!$ is a history expansion feature which expands to the last word of the previous command when used in a command line. So, basically ! is a history expansion character by default.</li></ul><p>Here are the examples you can try:</p><pre># Case 1<br>➜  ~ echo hello world<br>hello world<br>➜  ~ echo $_<br>world<br>➜  ~ echo !$<br>➜  ~ echo $_<br>world<br><br># Case 2<br>➜  ~ echo hello world<br>hello world<br>➜  ~ echo !$<br>➜  ~ echo world<br>world<br>➜  ~<br><br># playing around with history expansion (!)<br># recall 1st word<br>➜  ~ echo how are you<br>how are you<br>➜  ~ echo !!:1<br>➜  ~ echo how<br><br># recall 2nd word<br>➜  ~ echo how are you<br>how are you<br>➜  ~ echo !!:2<br>➜  ~ echo are<br><br># recall 3rd word<br>➜  ~ echo how are you<br>how are you<br>➜  ~ echo !!:3<br>➜  ~ echo you<br><br># recall last word (same as !$)<br>➜  ~ echo how are you<br>how are you<br>➜  ~ echo !!:$<br>➜  ~ echo you</pre><p>You will see when you enter !$ history expansion command, it was recalled and evaluated (Note: it was not executed).</p><p>I hope this might helpful for you too.</p><p>References:</p><ul><li><a href="https://www.gnu.org/software/bash/manual/html_node/Bash-Variables.html">https://www.gnu.org/software/bash/manual/html_node/Bash-Variables.html</a></li><li><a href="https://www.gnu.org/software/bash/manual/html_node/History-Interaction.html">https://www.gnu.org/software/bash/manual/html_node/History-Interaction.html</a></li><li><a href="https://www.gnu.org/software/bash/manual/html_node/Word-Designators.html">https://www.gnu.org/software/bash/manual/html_node/Word-Designators.html</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b53650e6e683" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>