Hi, I'm Yehuda Livne 👋
I build practical software products with a strong focus on AI workflows, browser applications, automation, and developer tools.
I enjoy taking an idea from zero to a working product: designing the architecture, implementing the system, solving unfamiliar engineering problems, and continuously improving the result.
- 🤖 AI agents and workflow automation
- 📄 Browser-first document tools (PDF processing, editing, compression)
- 🌐 Full-stack web applications
- 📱 Android applications
- ⚙️ Developer tools and engineering experiments
- 🎮 Browser-based retro gaming projects
A Hebrew-first AI productivity coach built with n8n, Gemini, and deterministic workflow state.
Highlights: LLM + workflow orchestration · File-backed memory · Telegram integration · Session lifecycle management · Timer engine · Prompt engineering
A browser-first PDF compression platform combining local processing, Web Workers, and optional server-side OCR.
Highlights: Web Workers · Next.js · TypeScript · pdf-lib · PDF.js · Background jobs · Performance optimization
A client-side PDF editor supporting annotations, search, history, and export.
Highlights: Canvas rendering · PDF.js · pdf-lib · IndexedDB · Session persistence · Undo/Redo architecture
A privacy-first browser PDF merger.
Highlights: Client-side processing · Localization · Offline support · Service Workers · Advanced PDF manipulation
🎮 C:\PLAY
A browser DOS game player powered by js-dos.
Highlights: Emulator integration · IndexedDB save states · GitHub Pages deployment · Static generation · Browser resource management
📱 KAN
An Android application that tracks screen usage and time away from the phone.
Highlights: Kotlin · Jetpack Compose · Foreground Services · StateFlow · Background lifecycle handling
Languages: JavaScript · TypeScript · Python · PHP · Kotlin · Bash
Frameworks & Platforms: React · Next.js · Android · Node.js · n8n
Infrastructure: AWS · Docker · Linux · GitHub Actions
AI: Google Gemini · LLM workflow orchestration · Prompt engineering · AI agent design
I like problems that combine multiple disciplines — AI systems, workflow automation, developer experience, system design, browser engineering, performance, state management, and human-computer interaction.
Most of my projects explore the same underlying tradeoff: where deterministic software ends, where AI begins, and how the two can work together reliably. I try to build products that are simple to use but technically interesting underneath.
Current focus: AI-powered developer tools, workflow automation, and browser-based productivity software.
Feel free to explore the repositories, open issues, or reach out if you'd like to discuss engineering, AI systems, or interesting technical problems.