DATA Agent Platform

Build, deploy, and scale production-ready AI agents

Features • Quick Start • Architecture • Deployment • Contributing

---

🚀 Overview

DATA is a production-grade LLM application development platform designed for teams building AI-powered products. From simple chatbots to complex multi-agent workflows, DATA provides the visual tools, infrastructure, and runtime to go from prototype to production in hours, not weeks.

What Makes DATA Different?

• No-Code to Pro-Code — Design workflows visually, extend with code when you need more power
• Production-Ready — Built-in observability, error handling, and performance monitoring
• Model Agnostic — 100+ LLM providers, embed, and rerank models out of the box
• Enterprise Secure — RBAC, SSO, audit logs, and data isolation

---

✨ Features

AI Agent Platform

Feature	Description
Visual Workflow Builder	Drag-and-drop pipeline design with real-time preview
Multi-Agent Orchestration	Chain, route, and parallel-execute AI agents
RAG Pipeline	Built-in document ingestion, chunking, embedding, and retrieval
50+ Built-in Tools	Web search, code execution, image generation, API connectors
Plugin Ecosystem	Install community plugins or build your own
Human-in-the-Loop	Pause workflows for human approval or input
Auto-Scaling	Horizontal scaling for production workloads

Model Management

• 100+ Providers: OpenAI, Anthropic, Google, AWS Bedrock, Azure, local LLMs (Ollama/Llama.cpp)
• Model Fallbacks: Define fallback chains for reliability
• Load Balancing: Distribute requests across model instances
• Usage Tracking: Monitor token consumption and costs

Observability

• Real-time Monitoring: Live workflow execution traces
• Logging: Structured logs with search and filtering
• Analytics: Usage patterns, response times, error rates
• Integration: OpenTelemetry, Langfuse, custom exporters

Developer Experience

• REST API: Full API access for custom integrations
• SDKs: Python, Node.js, and CLI toolkit
• DSL Export: Version-controlled app definitions
• CI/CD Ready: Docker-based deployment pipeline

---

🏗️ Architecture

┌─────────────────────────────────────────────────────────┐
│                     Web Frontend (Next.js)              │
│  ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐  │
│  │ Workflow │ │  Agent   │ │ Dataset  │ │ Settings │  │
│  │  Studio  │ │  Config  │ │ Manager  │ │   Page   │  │
│  └──────────┘ └──────────┘ └──────────┘ └──────────┘  │
└──────────────────────┬──────────────────────────────────┘
                       │ HTTP/WebSocket
┌──────────────────────▼──────────────────────────────────┐
│                    API Server (FastAPI)                  │
│  ┌─────────────┐  ┌────────────┐  ┌──────────────────┐ │
│  │  Workflow   │  │   Agent    │  │  Model Runtime   │ │
│  │   Engine    │  │  Runner    │  │    Manager       │ │
│  ├─────────────┤  ├────────────┤  ├──────────────────┤ │
│  │ RAG Service │  │ Tool System│  │ Plugin Registry  │ │
│  └─────────────┘  └────────────┘  └──────────────────┘ │
└──────────────────────┬──────────────────────────────────┘
                       │
┌──────────────────────▼──────────────────────────────────┐
│                    Data & Cache Layer                    │
│  ┌────────────┐ ┌────────────┐ ┌──────────────────┐    │
│  │ PostgreSQL │ │   Redis    │ │  Vector DB       │    │
│  │            │ │            │ │  (multi-engine)  │    │
│  └────────────┘ └────────────┘ └──────────────────┘    │
└──────────────────────────────────────────────────────────┘

Data Flow

User Request → API Gateway → Auth Check → Workflow Engine
    ↓
Agent Selection → Context Building → LLM Invocation
    ↓
Tools Execution ← RAG Retrieval ← Knowledge Base
    ↓
Response Assembly → Streaming → User

---

🚀 Quick Start

Prerequisites

• Docker & Docker Compose
• 4GB+ RAM recommended

One-Click Launch

# Clone the repository
git clone https://github.com/your-org/data-agent
cd data-agent

# Start all services
cd docker
docker compose up -d

# Access the platform
# Web UI:    http://localhost:3000
# API:       http://localhost:5001
# Swagger:   http://localhost:5001/docs

Manual Setup (Development)

# Backend
cd api
cp .env.example .env
uv sync
uv run flask db upgrade
uv run python app.py

# Frontend
cd web
cp .env.example .env.local
pnpm install
pnpm dev

---

🛠️ Configuration

Environment Variables

Variable	Default	Description
DATA_BIND_ADDRESS	0.0.0.0	API server bind address
DATA_PORT	5001	API server port
SECRET_KEY	(required)	App encryption key
DB_USERNAME	data	PostgreSQL username
DB_PASSWORD	data123456	PostgreSQL password
DB_HOST	localhost	PostgreSQL host
DB_PORT	5432	PostgreSQL port
REDIS_HOST	localhost	Redis host
REDIS_PORT	6379	Redis port

---

📚 Documentation

• User Guide — Getting started, tutorials, best practices
• API Reference — Full REST API documentation
• Plugin Development — Building custom plugins
• Contributor Guide — How to contribute

---

🔌 Plugins & Extensions

DATA supports a rich plugin ecosystem:

Type	Examples
Model Providers	OpenAI, Anthropic, Ollama, AWS Bedrock, Azure
Tools	Web search, Calculator, Code execution, Image gen
Vector Stores	Qdrant, Milvus, Weaviate, Pinecone
Document Loaders	PDF, HTML, Notion, Confluence, S3
Observability	Langfuse, OpenTelemetry, Sentry

---

🧪 Testing

# Backend tests
cd api
uv run pytest tests/unit_tests/ -v

# Frontend tests
cd web
pnpm test

# E2E tests
pnpm test:e2e

---

🤝 Contributing

We welcome contributions! See our Contributing Guide for details.

1. Fork the repo
2. Create your feature branch (git checkout -b feature/amazing)
3. Commit changes (git commit -m 'Add amazing feature')
4. Push to branch (git push origin feature/amazing)
5. Open a Pull Request

Code of Conduct

Please read our Code of Conduct.

---

📄 License

This project is licensed under the Apache License 2.0 — see the LICENSE file for details.

---

_{Built with ❤️ by the DATA Team}