{"slug": "how-i-built-a-3-3mb-native-windows-ai-desktop-app-with-deepseek-tailwind-goodbye", "title": "How I Built a 3.3MB Native Windows AI Desktop App with DeepSeek & Tailwind (Goodbye 150MB Electron Bloat)", "summary": "A developer built LeadHunter AI Desk, an open-source native Windows desktop application for B2B cold outreach that weighs just 3.3 MB and launches in under 0.2 seconds, avoiding the roughly 185 MB footprint of typical Electron-based tools. The app pairs a native Win32 core built with VisualNEO Win with a Tailwind CSS interface rendered through the pre-installed Microsoft Edge WebView2 runtime, and uses a bring-your-own-key model via OpenRouter with DeepSeek V3 to generate 1,000 personalized emails for about $0.04 in API credits. It sends mail through SMTP2GO's HTTPS REST API on port 443 to sidestep ISP blocks on traditional SMTP ports, and is released under the MIT License.", "body_md": "Every time I want to build a desktop utility in 2026, the modern web ecosystem gives me the same repetitive answer: *\"Just bundle it with Electron!\"*\n\nDon't get me wrong: Electron is an impressive engineering feat. But packaging an entire Chromium browser and Node.js runtime just to display a local database and fire off a few API calls means:\n\nA few weeks ago, I needed a specialized B2B cold outreach software for Windows — something that could import lead lists, craft hyper-personalized emails using modern LLMs (DeepSeek V3, Claude 3.5 Sonnet), and send them via transactional REST APIs without monthly SaaS fees.\n\nI refused to accept a 200MB bloated binary for this.\n\nInstead, I built **[LeadHunter AI Desk](https://github.com/sinlios/leadhunter-ai-desk)**: a 100% native Windows desktop application with a modern Tailwind CSS dark-mode UI that weighs **just 3.3 MB** in a portable package and launches in **0.18 seconds**.\n\nHere is how the architecture works and how you can build similar lightweight desktop tools.\n\nThe secret to building ultra-lightweight desktop applications without sacrificing modern HTML5/CSS aesthetics is simple: **stop distributing the browser runtime with your app**.\n\nEvery modern installation of Windows 10 and Windows 11 already includes the **Evergreen Microsoft Edge WebView2** runtime pre-installed and updated by Windows Update.\n\nTo orchestrate this, I used **[VisualNEO Win](https://visualneowin.com)** — a rapid application development (RAD) environment for Windows that compiles directly into native Win32 executables:\n\n```\n+---------------------------------------------------------+\n|                 LeadHunter AI Desk                      |\n+----------------------------+----------------------------+\n|   Native Win32 Core        |   Reactive UI Layer        |\n|   (VisualNEO Win Runtime)  |   (Edge WebView2 Runtime)  |\n+----------------------------+----------------------------+\n| * Zero Chromium bundle     | * Tailwind CSS Dark Mode   |\n| * Local INI key storage    | * Real-time KPI Dashboards |\n| * Native Windows Events    | * Interactive Data Tables  |\n| * CSV File Handlers        | * Batch Selection Modals   |\n+----------------------------+----------------------------+\n|        Bidirectional Async Messaging Bridge             |\n|        (wvPostMessage <--> OnWebMessageReceived)        |\n+---------------------------------------------------------+\n```\n\nBy decoupling the UI layer from a bundled browser:\n\n`.exe` is tiny.\n| Metric | LeadHunter AI Desk | Typical Electron Outreach Tool | \n|---|---|---|\n| **Download Size** | **3.3 MB** (Portable Zip) | 185 MB | \n| **Installed Footprint** | **~8 MB** | 420 MB | \n| **Startup Speed** | **< 0.2s (Instantaneous)** | 3.2 seconds | \n| **Idle RAM Footprint** | **~35 MB** | 580 MB | \n| **Pricing Model** | **$0 / Free & Open Source (BYOK)** | $49 to $99 / month | \n\nInstead of paying a $50/month SaaS subscription that marks up AI tokens by 500%, LeadHunter AI Desk uses a **BYOK (Bring Your Own Key)** model connected to **OpenRouter**:\n\n`{company_name}`, `{city}`, `{website}`, and `{my_service}`.\nWith **DeepSeek V3**, generating 1,000 unique, personalized cold emails costs approximately **$0.04 in direct API credits**. That represents a 99.8% cost reduction compared to commercial cloud outreach platforms.\n\nIf you have ever built a desktop tool that sends emails, you know the pain: residential ISPs and corporate networks routinely block outgoing traffic on **Port 25, 465, and 587** to prevent botnet spam.\n\nLeadHunter AI Desk circumvents this entirely by transmitting emails via **SMTP2GO's official HTTPS REST API (Port 443)**:\n\nThe entire project is open source under the **MIT License**.\n\nIf you want to inspect how the native Win32 runtime talks to the WebView2 frontend:\n\n```\n; Send data to the Tailwind UI frontend\nwvPostMessage \"WebBrowser1\" \"LOAD_CONFIG|{\\\"model\\\":\\\"deepseek/deepseek-chat\\\"}\"\n\n; Handle user interactions triggered from the HTML layer\n:OnWebMessageReceived\n  StrParse \"[WebMessage]\" \"|\" \"[Action]\" \"[Data]\"\n  If \"[Action]\" \"=\" \"START_OUTREACH\"\n    ...\n  EndIf\nReturn\n```\n\nYou can download the full project, edit the `.pub` file, and test it yourself:\n\nThe web is an amazing place, and HTML/Tailwind CSS is still the most expressive UI styling system in existence. But we don't need to ship a 150MB operating system emulator just to draw a few buttons and tables on Windows.\n\nWhat is your preferred approach to building desktop software in 2026? Are you sticking with Electron, experimenting with Tauri, or rediscovering native Windows tooling?\n\nI'd love to hear your thoughts in the comments below! ⭐", "url": "https://wpnews.pro/news/how-i-built-a-3-3mb-native-windows-ai-desktop-app-with-deepseek-tailwind-goodbye", "canonical_source": "https://dev.to/sinlios/how-i-built-a-33mb-native-windows-ai-desktop-app-with-deepseek-tailwind-goodbye-150mb-electron-25c7", "published_at": "2026-09-10 11:14:43+00:00", "updated_at": "2026-09-10 11:28:26.183613+00:00", "lang": "en", "topics": ["ai-tools", "ai-products", "developer-tools", "generative-ai"], "entities": ["LeadHunter AI Desk", "VisualNEO Win", "Microsoft Edge WebView2", "DeepSeek V3", "OpenRouter", "SMTP2GO", "Claude 3.5 Sonnet", "Windows"], "alternates": {"html": "https://wpnews.pro/news/how-i-built-a-3-3mb-native-windows-ai-desktop-app-with-deepseek-tailwind-goodbye", "markdown": "https://wpnews.pro/news/how-i-built-a-3-3mb-native-windows-ai-desktop-app-with-deepseek-tailwind-goodbye.md", "text": "https://wpnews.pro/news/how-i-built-a-3-3mb-native-windows-ai-desktop-app-with-deepseek-tailwind-goodbye.txt", "jsonld": "https://wpnews.pro/news/how-i-built-a-3-3mb-native-windows-ai-desktop-app-with-deepseek-tailwind-goodbye.jsonld"}}