Skip to main content

​
What is LoopOS AI?

LoopOS AI is a standalone product that can be used headlessly or integrated within LoopOS applications. It provides a robust set of AI-powered services designed to automate and enhance operations in circular economy marketplaces.

Multi-Agent Architecture

Specialized AI agents work together to handle complex workflows like product identification, validation, and pricing.

API-First Design

All services are accessible via REST API, making integration seamless across different platforms and applications.

Context-Aware

Services leverage business context and historical data to provide intelligent, relevant responses.

Observable & Monitored

Built-in observability with Langfuse integration for tracing, token tracking, and cost monitoring.

​
Core Principles

LoopOS AI is built on several key principles:
  • Technological isolation: Services are self-contained and can operate independently
  • Leverage existing tools: Built on proven frameworks like OpenAI Agents and FastAPI
  • Super agile development: Fast deploy cycles with script-based development
  • Production testing: MVPs are tested in production with aggressive monitoring
  • Continuous improvement: Focus on improving models rather than over-customizing current implementations

​
Key Components

​
LoopOS AI Server

The Python-based server hosts all AI agents and services, serving as the backbone of the LoopOS AI ecosystem. It features:
  • Multi-agent architecture with specialized agents
  • Built-in guardrails for safety and compliance
  • Scalable and modular infrastructure
  • Centralized prompt management
  • Performance and cost monitoring

​
LoopOS Context Provider

Context is everything for LoopOS AI. All context must be managed and provided by LoopOS applications, particularly LoopOS Core. The context includes:
  • Business operations information
  • Use cases and models
  • Item data and metadata
  • Historical context for comprehensive understanding

​
Use Cases

Product Submission

Conversational AI agent guides users through product submission, identifying items from catalog and collecting required information.

Marketplace Listings

Generate optimized marketplace listings with pricing, descriptions, and condition assessment for C2C transactions.

Content Validation

Automatically validate marketplace listings, brands, and products against quality standards and catalog references.

Value Estimation

Estimate item values using web search and market analysis for Portuguese marketplaces.

ID Validation

Extract and validate information from ID documents (passports, ID cards) with OCR and validation checks.

Protocol Generation

Automatically generate structured protocols (question sets) for collecting item information by category.

​
Getting Started

1

Explore the Services

Review the available services in the Services section to understand what LoopOS AI can do.
2

Understand the Architecture

Learn about the Architecture to understand how services are structured and how agents work together.
3

Try the Quickstart

Follow the Quickstart Guide to make your first API call and see LoopOS AI in action.
4

Integrate with Your App

Check the AI Hub for integration guides, prompt examples, and webhook configurations.

​
Next Steps