<?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>Forem: Dicon | H.I. MET ARCHITECT</title>
    <description>The latest articles on Forem by Dicon | H.I. MET ARCHITECT (@met_architect).</description>
    <link>https://forem.com/met_architect</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.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3861558%2F550f2dcb-3017-41bc-a626-453c502d46e3.jpg</url>
      <title>Forem: Dicon | H.I. MET ARCHITECT</title>
      <link>https://forem.com/met_architect</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://forem.com/feed/met_architect"/>
    <language>en</language>
    <item>
      <title>I Brought My Local-First "Obsidian-to-AI" Bridge to Android (With Massive Pro Upgrades)</title>
      <dc:creator>Dicon | H.I. MET ARCHITECT</dc:creator>
      <pubDate>Mon, 06 Apr 2026 13:12:19 +0000</pubDate>
      <link>https://forem.com/met_architect/i-brought-my-local-first-obsidian-to-ai-bridge-to-android-with-massive-pro-upgrades-1h12</link>
      <guid>https://forem.com/met_architect/i-brought-my-local-first-obsidian-to-ai-bridge-to-android-with-massive-pro-upgrades-1h12</guid>
      <description>&lt;p&gt;A while ago, I shared my journey of building Second Brain Joiner—a local-first iOS app designed to safely merge Obsidian markdown vaults and feed them into LLMs like ChatGPT and Claude without relying on cloud servers or risking privacy.&lt;/p&gt;

&lt;p&gt;The response from the community was amazing. But the number one request I received was loud and clear: "When is the Android version coming?"&lt;/p&gt;

&lt;p&gt;I hear you. As an indie developer and a massive Obsidian fan myself, I knew I had to make it happen. Today, I’m incredibly excited to announce that Second Brain Joiner is officially live on Google Play! 🎉&lt;br&gt;
　&lt;br&gt;
　&lt;br&gt;
　&lt;br&gt;
👇 [Get it on Google Play]&lt;br&gt;
&lt;a href="https://play.google.com/store/apps/details?id=com.himetarchitect.secondbrainjoiner" rel="noopener noreferrer"&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%2Fhrch574bszl5pnrghm8h.png" alt="Get it on Google Play" width="646" height="250"&gt;&lt;/a&gt;&lt;br&gt;
　&lt;br&gt;
　&lt;br&gt;
🎥 [Watch the Quick Demo]&lt;br&gt;
  &lt;iframe src="https://www.youtube.com/embed/Boc3UqoohaE"&gt;
  &lt;/iframe&gt;
&lt;br&gt;
　&lt;br&gt;
　&lt;/p&gt;

&lt;h2&gt;
  
  
  🚀 Building the Android Version:
&lt;/h2&gt;

&lt;p&gt;Not Just a Port, an Evolution&lt;br&gt;
When I started building the Android version, I didn't want to just port the iOS app as-is. As I used the app daily for my own workflow, I realized that power users need more granular control over what they send to the AI.&lt;/p&gt;

&lt;p&gt;Throwing your entire vault at an LLM is a bad idea (hello, context window limits and hallucination!). You need precise context.&lt;/p&gt;

&lt;p&gt;So, I built the Android version from the ground up with a completely new set of "Pro" features tailored for heavy AI users:&lt;br&gt;
　&lt;br&gt;
　&lt;/p&gt;

&lt;h2&gt;
  
  
  1. OpenAI API-Based Token Calculation
&lt;/h2&gt;

&lt;p&gt;Word counts are useless for AI. You need to know the exact token count. I implemented a highly accurate token calculation system based on the OpenAI API standard, so you know exactly how heavy your prompt is before you paste it.&lt;br&gt;
　&lt;br&gt;
　&lt;/p&gt;

&lt;h2&gt;
  
  
  2. Advanced Filtering (AND/OR Logic)
&lt;/h2&gt;

&lt;p&gt;You can now pinpoint exactly which notes to merge. The app allows you to combine Tags and Last Modified Dates (From-To) using flexible AND/OR conditions. Want to extract only notes tagged with #project-x that were updated in the last 7 days? Now you can do it in seconds.&lt;br&gt;
　&lt;br&gt;
　&lt;/p&gt;

&lt;h2&gt;
  
  
  3. Auto Chunking for Large Vaults
&lt;/h2&gt;

&lt;p&gt;If your merged text exceeds practical LLM limits, the app now features an Auto Chunking system. It intelligently splits your massive markdown outputs into manageable chunks so you can feed them to ChatGPT or Claude sequentially without breaking the context window.&lt;br&gt;
　&lt;br&gt;
　&lt;/p&gt;

&lt;h2&gt;
  
  
  🍎 What about iOS users? (The Backport Promise)
&lt;/h2&gt;

&lt;p&gt;If you are currently using the iOS version, you might be thinking, "Hey, what about us?" Don't worry! I haven't forgotten you. Developing these advanced features on Android first allowed me to refine the logic and UI. I am officially announcing that all of these new features (Advanced Filtering, Auto Chunking, and Token Calculation) will be backported to the iOS version in an upcoming major update. Stay tuned!&lt;br&gt;
　&lt;br&gt;
　&lt;/p&gt;

&lt;h2&gt;
  
  
  💡 Fully Local, Fully Secure
&lt;/h2&gt;

&lt;p&gt;Just like the iOS version, the Android app respects your privacy. It processes your .md files entirely locally on your device. No servers, no data harvesting, no sneaky API calls in the background. Your second brain stays yours.&lt;br&gt;
　&lt;br&gt;
　&lt;br&gt;
If you are an Android user looking to supercharge your Obsidian + AI workflow, I would love for you to give it a try and let me know what you think!&lt;br&gt;
　&lt;br&gt;
Happy Linking! 🧠✨&lt;/p&gt;

</description>
      <category>ai</category>
      <category>android</category>
      <category>productivity</category>
      <category>showdev</category>
    </item>
    <item>
      <title>I built a local-first Obsidian suite to safely feed my vault to AI 🛠️🐕</title>
      <dc:creator>Dicon | H.I. MET ARCHITECT</dc:creator>
      <pubDate>Sat, 04 Apr 2026 22:20:23 +0000</pubDate>
      <link>https://forem.com/met_architect/i-built-a-local-first-obsidian-suite-to-safely-feed-my-vault-to-ai-9c9</link>
      <guid>https://forem.com/met_architect/i-built-a-local-first-obsidian-suite-to-safely-feed-my-vault-to-ai-9c9</guid>
      <description>&lt;p&gt;Hi everyone! 👋 I’m Dicon, a Hospital IT Architect and Indie Dev from Japan. &lt;br&gt;
And this is my chief assistant/QA tester, Maru the Shiba.🐕🌸&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%2F0rddbyx5lm9pi9fw7iul.jpg" 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%2F0rddbyx5lm9pi9fw7iul.jpg" alt="Maru the Shiba" width="800" height="1067"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Working in medical IT, I know how critical data privacy is. I also happen to be a massive Obsidian user with a highly complex vault. &lt;/p&gt;

&lt;p&gt;Recently, I wanted to leverage AI to analyze my notes, but I hit a wall: &lt;strong&gt;I didn't want to upload my entire personal vault to the cloud, and preparing the exact context for AI was a nightmare.&lt;/strong&gt; Also, exporting complex Markdown to beautiful PDFs was always broken.&lt;/p&gt;

&lt;p&gt;So, I decided to build my own solution.&lt;br&gt;
　&lt;br&gt;
　&lt;/p&gt;

&lt;h2&gt;
  
  
  🛠️ The Solution: A Local-First Input/Output Suite
&lt;/h2&gt;

&lt;p&gt;I built two local-first apps to solve the input/output problems for Obsidian power users.&lt;br&gt;
　&lt;/p&gt;

&lt;h3&gt;
  
  
  1. Second Brain Joiner (SBJ) - The Input Engine
&lt;/h3&gt;

&lt;p&gt;SBJ allows you to merge, and copy your specific Obsidian notes perfectly formatted for AI context. All processing is done &lt;strong&gt;locally&lt;/strong&gt; on your device. No cloud risks.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://youtube.com/shorts/YVI3L7hTjpA" rel="noopener noreferrer"&gt;Second Brain Joiner-Demo&lt;/a&gt;&lt;br&gt;
  &lt;iframe src="https://www.youtube.com/embed/YVI3L7hTjpA"&gt;
  &lt;/iframe&gt;
　&lt;br&gt;
　&lt;/p&gt;

&lt;h3&gt;
  
  
  2. VaultPress (VP) - The Output Engine
&lt;/h3&gt;

&lt;p&gt;What about output? I built VaultPress, a powerful local rendering engine that turns your complex Markdown notes into flawless PDFs instantly. Beautiful input deserves beautiful output.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://youtube.com/shorts/vFAoHL3SlrA" rel="noopener noreferrer"&gt;VaultPress-Demo&lt;/a&gt;&lt;br&gt;
  &lt;iframe src="https://www.youtube.com/embed/vFAoHL3SlrA"&gt;
  &lt;/iframe&gt;
&lt;br&gt;
　　&lt;br&gt;
　&lt;/p&gt;

&lt;h2&gt;
  
  
  🚫 The Best Part: No Subscriptions
&lt;/h2&gt;

&lt;p&gt;I hate monthly fees as much as you do. &lt;br&gt;
As an indie dev, I want to provide tools that you truly own. Both a&lt;/p&gt;

&lt;p&gt;pps are &lt;strong&gt;one-time purchases&lt;/strong&gt;. Buy once, use forever.&lt;br&gt;
　&lt;br&gt;
　&lt;/p&gt;

&lt;p&gt;🍏 &lt;strong&gt;For iOS Users:&lt;/strong&gt;&lt;br&gt;
The complete Input/Output suite is available NOW on the App Store as an App Bundle!&lt;br&gt;
　&lt;br&gt;
&lt;a href="https://apps.apple.com/us/app-bundle/second-brain-pro-bundle/id1887592238" rel="noopener noreferrer"&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%2F6ilt79dr3besj2743ibo.png" alt="logo" width="150" height="50"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;　&lt;br&gt;
🤖 &lt;strong&gt;For Android Users:&lt;/strong&gt;&lt;br&gt;
Get ready! SBJ (with massive new features) and VP are currently in review. I will announce the release very soon!&lt;/p&gt;




&lt;p&gt;I’d love to hear your thoughts and feedback. How do you currently handle exporting and feeding your Obsidian notes to AI? Let me know in the comments! &lt;/p&gt;

&lt;p&gt;Cheers from Japan! 🇯🇵&lt;/p&gt;

</description>
      <category>showcase</category>
      <category>obsidian</category>
      <category>productivity</category>
      <category>ai</category>
    </item>
  </channel>
</rss>
