Projekt is a visual workspace designed for builders that wraps AI coding agents in a real-time preview, editable DOM, and prompt-first interface. It enables users to run multiple AI coding agents side-by-side while seeing instant visual feedback as code is generated, all within a single unified environment.
Traditional development workflows require constant switching between terminals, code editors, and browser previews, creating friction and slowing down the building process. Projekt solves this by providing an integrated workspace where prompting, previewing, and refining happen simultaneously, eliminating context switching and accelerating the journey from idea to shipped product.
Projekt allows users to connect and run multiple AI coding agents simultaneously, including Claude Code, Codex, Gemini CLI, and Opencode. These agents work side-by-side in the same workspace, with real-time visibility into what each agent is doing as they generate code, analyze project structures, and create components.
The workspace features a live preview panel that updates in real-time as AI agents write code. Users can describe what they want through natural language prompts and immediately see the visual results, enabling rapid iteration and visual feedback loops without manual browser refreshing or build steps.
Every element in the preview is inspectable, selectable, and editable via direct manipulation. Users can click on any visual element to edit it inline, tweak styles visually, and make adjustments without digging through code files, combining the best of both prompt-based generation and visual editing.
Projekt operates as a complete workflow in one window, integrating file explorer, live preview, and chat interfaces without requiring users to switch between terminal, browser, and editor applications. This unified approach keeps the entire building process contained within a single workspace.
Users benefit from dramatically reduced context switching, faster iteration cycles through immediate visual feedback, and complete control over their AI infrastructure. The workspace enables builders to go from idea to shipped product more efficiently by streamlining the entire development workflow into one cohesive environment.
admin
Concrete use cases include building landing pages with hero sections and waitlist forms, creating dashboard interfaces, developing design systems, and prototyping web applications. Users can open any existing project folder and immediately start building with AI assistance, whether creating new components or modifying existing ones.
Projekt targets builders who go from idea to shipped product without waiting on anyone—essentially assuming the user is the whole team. It integrates with users' existing API keys for AI models, supports any project structure without proprietary formats, and currently offers a macOS desktop application available for download.
Projekt represents a fundamentally different approach to AI-assisted development by being AI-native from the ground up rather than adding AI features to existing tools, creating a workspace specifically optimized for the prompt-preview-refine workflow that modern builders need.
Projekt targets builders who go from idea to shipped product without waiting on anyone—essentially assuming the user is the whole team. These are designers who engineer, developers who build complete products, and creators who work independently from concept to deployment. The workspace is designed for people who currently switch between terminals, code editors, and browser previews and want an integrated solution that streamlines their entire workflow. It serves users who want to leverage multiple AI coding agents simultaneously while maintaining control over their API keys and avoiding vendor lock-in.