For years, I searched for the right programmer’s editor for my daily work. I tried countless tools along the way, but they always fell into the same traps: either they were bloated, sluggish, and consumed gigabytes of RAM, or they were locked behind absurdly expensive enterprise licenses. In the end, I decided to build VIDE — an editor focused purely on speed, control, and zero friction for the developer.

Along the way, I integrated practical AI capabilities powered by Google’s cost-effective models. To be clear: VIDE isn’t trying to compete with massive autonomous coding agents like Antigravity, Claude Code, or OpenAI’s developer agents. Instead, the AI serves as a fast completion assistant combined with built-in code analysis tools — making it a remarkably powerful aid for exploring, dissecting, and understanding complex codebases without burning through expensive tokens or getting in your way.

Planning a complex AI initiative or high-scale platform?

VIDE showcases my architectural philosophy: lean, high-performance systems with zero bloat and disciplined engineering. Beyond tooling, I work as a senior technical advisor and systems architect, helping companies design and scale advanced AI systems, complex e-commerce platforms, and mission-critical software.

Contact: Ori Kuttner — ori@heliconbooks.com

Vector IDE (vide) is an ultra-lightweight, lightning-fast native integrated development environment designed for high-performance coding without the overhead of Electron. Built with Tauri v2 (Rust backend), Angular 21 (Signals-driven frontend), and Microsoft's Monaco Editor, Vector IDE provides minimal memory consumption (~40–80 MB RAM), and deep AI assistance powered directly by Google Gemini.

Blazing Fast & Low Memory Footprint: Native desktop application powered by Rust and Tauri v2. Consumes a fraction of the resources required by Electron-based editors.

Full Monaco Code Editor Integration:

Multi-cursor editing, bracket matching, code folding, mini-map, and rich language syntax highlighting.

Dedicated web workers for TypeScript/JavaScript, HTML, CSS/SCSS, and JSON.

Custom OmniFlux language support with Monarch tokenization and snippets registry.

Direct Google Gemini AI Integration:

In-Editor Chat Assistant: Real-time chat aware of your active file, buffer content, and exact cursor coordinates.

Workspace Tool Calling:

list_files: Discover project structure within workspace boundaries.

read_file: Inspect project code safely without leaving the editor.

replace_text: Guarded, unique-target code modifications strictly gated by explicit user confirmation.

open_file: Navigate to target files across your workspace.

set_cursor: Jump directly to specific line and column positions.

highlight_lines: Open any project file, make it active, and visually highlight execution flows and line spans in Monaco.

Dual-Model Architecture: Configure independent models for inline code completions (e.g. gemini-3.5-flash-lite) and conversational reasoning (e.g. gemini-3.8-flash).

Bilingual & Directional Support: Seamless LTR / RTL toggle with automatic input language detection (dir="auto").

Single-Instance CLI & IPC:

Open or switch workspaces directly from the command line: vide [path].

External tool navigation: vide <workspace> <file> --highlight <start>-<end> focuses existing running instances and highlights targeted line spans instantly.

Git Integration: Visual branch indicator, file change tracking, and integrated diff comparison.

Workspace File Management: Fast tree explorer, multi-tab buffer management, fuzzy search, and customizable themes.

Screenshots

Main Workspace & Editor

┌────────────────────────────────────────────────────────┐ │ Vector IDE (vide) │ │ │ │ ┌──────────────────────────────────────────────────┐ │ │ │ Angular 21 Frontend │ │ │ │ • Monaco Editor & Workers (TS, HTML, CSS, JSON)│ │ │ │ • AI Chat Assistant & Floating Trigger │ │ │ │ • File Tree Explorer, Tabs & Status Bar │ │ │ │ • Google Gemini v1beta REST Client │ │ │ └────────────────────────┬─────────────────────────┘ │ │ │ IPC (Tauri Core) │ │ ┌────────────────────────┴─────────────────────────┐ │ │ │ Rust Tauri Backend │ │ │ │ • Native File System & Git Process Execution │ │ │ │ • Single-Instance Workspace Socket (IPC) │ │ │ │ • Window & Webview Lifecycle Management │ │ │ └──────────────────────────────────────────────────┘ │ └────────────────────────────────────────────────────────┘

Getting Started

Prerequisites

Ensure you have the following installed on your system:

sudo apt-get update sudo apt-get install -y libwebkit2gtk-4.1-dev build-essential curl wget file libxdo-dev libssl-dev libayatana-appindicator3-dev librsvg2-dev

macOS: Xcode Command Line Tools (xcode-select --install).

Windows: Microsoft Visual Studio C++ Build Tools and WebView2.

Installation

Clone the repository and install npm packages:

git clone https://github.com/OriKuttner/vide.git cd vide npm install

Running in Development

To start the local Angular development server and launch the Tauri desktop application:

npx tauri dev

Building for Production

To compile both frontend and native desktop release binaries:

npx tauri build

The resulting packages (.deb, .AppImage, .dmg, or .msi / .exe) will be output to src-tauri/target/release/bundle/.

Vector IDE includes a powerful CLI parser with single-instance reuse:

Open current directory as workspace

vide .

Open a specific project folder

vide /path/to/project

Navigate to a specific file and highlight a line range in a running instance

vide /path/to/project src/main.rs --highlight 42-65

Configuring AI Assistant

Obtain a Google Gemini API key from Google AI Studio.

In Vector IDE, open Settings (Gear icon in the top header or click the settings icon inside the AI Chat window).

Paste your Gemini API key in the AI Chat Assistant section.

Select your preferred reasoning model (default: gemini-3.8-flash) and system prompt.

AI Assistant Scope & Purpose

Note on Vector AI Scope: Vector IDE includes an integrated coding assistant specifically tailored for in-editor software analysis, architecture comprehension, and interactive code flow tracing.

Vector AI is designed to help developers navigate complex logic, trace execution paths across files, pinpoint bugs, and understand system architecture using real-time editor line highlighting and file context.

Under no circumstances is it intended to replace autonomous agent development environments or full CLI coding tools such as Google AntiGravity, Anthropic Claude Code, or OpenAI Canvas / Operator / Cursor. Its strength lies in being a fast, lightweight, and disciplined companion inside your native desktop editor.

Shortcut Action

Ctrl + S / Cmd + S Save current file

Ctrl + P / Cmd + P Quick Open / File Search

Ctrl + F / Cmd + F Find in active editor

Ctrl + H / Cmd + H Find and Replace

`Ctrl + `` Toggle Terminal / Logs

Ctrl + = / Ctrl + - Zoom In / Zoom Out

Ctrl + 0 Reset Zoom

Advisory & Architecture Consulting

VIDE is proof of disciplined, zero-bloat engineering. Building complex systems correctly requires deep architectural clarity before writing a single line of application code.

I advise CTOs, founders, and engineering leaders on designing resilient, cost-effective architectures and navigating technical challenges:

Looking for an experienced eye on your architecture? Reach out to Ori Kuttner at ori@heliconbooks.com.

Contributions are welcome! Please follow these guidelines:

Vector IDE (vide) is distributed under a Dual Licensing model:

For enterprise inquiries, dedicated proprietary builds, custom feature development, or dual-license exemptions, please contact: Ori Kuttner: ori@heliconbooks.com