Skip to main content

We use cookies for analytics. Privacy

Back to Work
AI & Machine LearningCase study

Oxide Studio

Universal AI Content Production Engine

Project Focus
DioxusRustTailwind CSSAI Model GatewaySQLiteIndexedDBCustom Model Training
Oxide Studio
Hundreds of unified models
AI Models
Web, Desktop, Mobile
Platforms
Dioxus (Rust)
Framework
SQLite + IndexedDB
Storage
01

Challenge

Managing hundreds of AI models from different providers (image, video, audio, text) requires complex API abstraction. Making this accessible to non-technical users while maintaining power for professionals is a design challenge. The UI must adapt contextually from chat to complex workflow editing.

02

Solution

Oxide Studio uses Dioxus to compile Rust to WASM and ship one codebase across platforms. The Recipe Engine breaks each job into ordered steps. A Brand Brain system trains custom models on brand assets. A unified API layer routes requests to every AI model through one path.

03

Results

  • Cross-platform (Web, Desktop, Mobile) via Rust/WASM
  • Hundreds of AI models unified through single interface
  • Recipe Builder with intuitive step types
  • Brand Brain custom training system
  • Chat-to-recipe conversational AI
  • Local-first with SQLite/IndexedDB persistence

System Architecture

AI content production with conversational workflow creation and unified model access

frontend
backend
database
service
external
Goal-based promptsWorkflow editsBrand assetsModel requestsSave outputHistory display
Dioxus UI
Cross-platform interface
Chat Panel
Conversational AI
Context Panel
Recipe Builder
Recipe Engine
Workflow orchestration
Brand Brain
Custom training
Generation Library
History and versions
AI Model Gateway
Unified model access

AI content production with conversational workflow creation and unified model access

Facing Similar Challenges?

Every business is different, but the problems tend to rhyme. If someone sent you, get in touch and tell us about yours.

A conversation, not a pitch
No obligation
We reply when we can