cd /news/artificial-intelligence/how-i-built-4-custom-apps-in-one-wee… · home topics artificial-intelligence article
[ARTICLE · art-34030] src=dev.to ↗ pub= topic=artificial-intelligence verified=true sentiment=↑ positive

How I Built 4 Custom Apps in One Week to Solve My Real-Life Problems (With an AI Agent)

A developer built four custom applications in one week using an autonomous AI agent called Hermes to solve personal problems. The apps include a remote terminal with viewport rotation, a meditation PWA, a gym tracker, and a food diary with Gemini Vision AI integration.

read4 min views2 publishedJun 19, 2026

We are always told that programming is a tool for problem-solving. But this week, I experienced this truth on a completely different level. Using my autonomous AI Agent (Hermes), software development became my real-time, daily survival tool.

Instead of searching for half-baked third-party apps, dealing with intrusive ads, or compromising on terrible mobile UX, I spent the last few days building four highly customized, production-grade applications to solve exact friction points in my daily life and for my family.

Here is the story of how these four apps came to life.

  • Remote Terminal (Defeating NoMachine's Screen Rotation Limit) The Problem: I use NoMachine to control my main development PC remotely from my mobile phone. However, I ran into an incredibly frustrating limitation: there was absolutely no way to rotate the remote display viewport by 90 degrees to fit my mobile screen orientation properly.

The Solution: Instead of giving up or hunting for obscure system-level workarounds, I built Remote Terminal. It is a lightweight remote execution client designed with a single killer feature: a built-in, hardware-accelerated 90-degree viewport rotation. Now, I can access and control my remote terminal rig perfectly, at the exact angle I need.

  • Tefila Meditation (From Print Draft to Interactive PWA) The Problem: I saw the draft of my wife's new meditation and prayer book right before it was sent to the physical print shop. I instantly realized that this content deserved an interactive, accessible digital life.

The Solution: Within 24 hours, we built a generic, highly optimized book framework and compiled it as a Flutter Web PWA and Native Android App.

  • Autoplay-Safe Audio: Implemented a synchronous audio player engine to bypass aggressive mobile browser autoplay blocks.

  • Beautiful Visuals: Added native vector-animated CustomPaint widgets (a flickering flame, flowing water, and rotating mandalas).

  • Offline-First Premium Model: Built a secure offline licensing system that unlocks chapters via secret coupon codes.

  • Live Proofreading Portal: Included a hidden, secure proofreader screen (edit2026) so she can edit chapters and export updated Dart code directly from her phone!

  • Deployed live on GitHub Pages: Check it out here.

  • My Personal Gym Companion The Problem: Most fitness and gym trackers on the app store are bloated with subscription popups, social feeds, and annoying ads. I just wanted a clean, fast way to log my workout routines and track my personal metrics while at the gym.

The Solution: While heading out to my workout session, I scaffolded a lightweight, completely local Gym Companion app. It is entirely ad-free, runs completely offline, and lets me track my progress, sets, and health metrics with zero distraction.

  • Today: Visual Food Tracker (With Gemini Vision AI Integration) The Problem: Today, I visited my diabetes specialist. She requested that I start keeping a meticulous daily food diary to monitor my dietary habits. The thought of manually typing every meal, ingredient, and portion size into a clunky text-based app sounded exhausting.

The Solution: I went home and, within a few hours, we built Food Tracker. It is an offline-first, visual-centric tracking app. I just released version 3.0:

  • Multi-Image Support: You can snap multiple photos per meal to track individual food items separately.
  • Smooth Carousel UI: On the home screen, meal cards turn into interactive swipeable carousels (PageView) so you can flip through the photos of your meal.
  • Google Gemini Vision AI Integration: I integrated Google's official google_generative_ai SDK (using Gemini Flash). You take a photo, tap the "🪄 Smart AI Analysis" button, and the AI:
  • Identifies the food items and auto-updates the captions for each image.
  • Calculates estimated nutritional values (calories, protein, carbs, and fats).
  • Automatically writes a detailed nutritional summary in the notes and suggests a cohesive title for the entire meal in Hebrew.

The Takeaway: Software as a Personal Utility

We are living in an era where the barrier between a raw idea and a working, compiled application has completely evaporated. When you pair your domain knowledge with a highly capable, autonomous AI agent, you stop being a mere consumer of software.

You become a producer of your own tools, tailoring technology to fit your life, your health, and your family's needs in real-time.

What is the next custom app you are going to build for yourself? Let me know in the comments!

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @hermes 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/how-i-built-4-custom…] indexed:0 read:4min 2026-06-19 ·