Skip to content

EfeDurmaz16/aspendos-deploy

Repository files navigation

YULA OS

YULA OS

Your Universal Learning Assistant

An AI-native operating system for thought — with persistent memory, agentic RAG, and proactive intelligence.

Live Demo TypeScript Next.js Vercel AI SDK


Overview

YULA isn't just another chatbot. It's an AI operating system that remembers everything, learns your patterns, and proactively helps you before you even ask.

┌─────────────────────────────────────────────────────────────┐
│                                                             │
│   "Remember that restaurant I mentioned last month?"        │
│                                                             │
│   YULA instantly retrieves the context, knows your          │
│   preferences, and suggests a reservation.                  │
│                                                             │
└─────────────────────────────────────────────────────────────┘

Key Features

Agentic RAG Router

Intelligent routing that decides when to search your memory — not just how.

// Fast route: Pattern matching (no LLM call, <1ms)
"Hi"  direct_reply

// LLM route: Complex decisions via Groq Llama (<100ms)
"What did we discuss about React hooks?"  rag_search

Persistent Memory

Every conversation becomes searchable knowledge. Vector embeddings via Qdrant enable semantic search across your entire history.

PAC Notifications

Proactive AI Callbacks — YULA doesn't just respond, it reaches out when it matters.

┌────────────────────────────────────────┐
│  🔔 PAC Notification                   │
│                                        │
│  "You mentioned wanting to review      │
│   your project proposal today.         │
│   Ready when you are."                 │
│                                        │
└────────────────────────────────────────┘

Multi-Provider AI

Unified interface to the best models, each chosen for their strengths:

Provider Model Use Case
Groq Llama 3.1 Fast routing decisions
Anthropic Claude 3.5 Complex reasoning & code
OpenAI GPT-4o General assistance

Architecture

yula/
├── apps/
│   ├── web/                 # Next.js 15 + React 19
│   └── mobile/              # React Native (Expo)
├── services/
│   └── api/                 # Hono API server
├── packages/
│   └── db/                  # Prisma + PostgreSQL
└── lib/
    └── ai/                  # Vercel AI SDK layer
        ├── router.ts        # Agentic RAG router
        ├── providers.ts     # Multi-provider registry
        └── generate.ts      # Unified generation

Tech Stack

Frontend

  • Next.js 15 (App Router)
  • React 19
  • Tailwind CSS 4
  • Framer Motion
  • shadcn/ui + Radix

Backend

  • Hono (Bun runtime)
  • PostgreSQL + Prisma
  • Qdrant (Vector DB)
  • Better Auth

AI Layer

  • Vercel AI SDK
  • @ai-sdk/openai
  • @ai-sdk/anthropic
  • @ai-sdk/groq

Infrastructure

  • Vercel (Web)
  • Railway (API + DB)
  • GitHub Actions CI/CD

Quick Start

# Clone
git clone https://github.com/EfeDurmaz16/aspendos-deploy.git
cd aspendos-deploy

# Install
bun install

# Setup database
bun db:generate
bun db:push

# Environment
cp .env.example .env.local
# Add your API keys

# Run
bun dev

Environment Variables

# AI Providers
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...
GROQ_API_KEY=gsk_...

# Database
DATABASE_URL=postgresql://...

# Vector Store
QDRANT_URL=https://...
QDRANT_API_KEY=...

# Auth
BETTER_AUTH_SECRET=...
BETTER_AUTH_URL=https://yula.dev

Design Philosophy

The Monolith Aesthetic

YULA embraces a bold, minimal design language:

Element Value
Background #050505 / #000000
Accent Electric Amber #F59E0B
Border #FFFFFF10
Font Geist / Geist Mono

No purple-blue gradients. No visual noise. Just signal.


API Reference

Chat Stream

POST /api/chat/stream
Content-Type: application/json

{
  "messages": [{ "role": "user", "content": "Hello" }],
  "model": "gpt-4o"
}

Memory Search

POST /api/memory/search
Content-Type: application/json

{
  "query": "React hooks discussion",
  "limit": 5
}

Health Check

GET /api/health

→ { "status": "healthy", "uptime": 12345 }

Contributing

We welcome contributions! Please see DEPLOYMENT.md for setup instructions.

# Run tests
bun test

# Type check
bun run typecheck

# Lint
bun biome check .

License

MIT © 2024 YULA OS


yula.dev · Built with Vercel AI SDK

Your thoughts, amplified.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors