Cart·AI Product Translator

Cart·AI Product Translator is a high-performance, asynchronous microservice written in Go (Golang). It is a core component of the broader Cart·AI Ecosystem, designed to work in tandem with the primary Java Spring Boot backend and the AI-driven Python worker.

🔗 Ecosystem Integration

This service operates as a decoupled enrichment step in the Cart·AI event-driven architecture:

1. The Cart·AI Backend publishes a translation.requested event to Apache Kafka when a product requires internationalization.
2. This Product Translator service consumes the event asynchronously.
3. It performs the translation using the Google Cloud Translation API.
4. Finally, it publishes a translation.completed event back to Kafka, which the main backend consumes to update the catalog.

🏛️ System Architecture

This project strictly follows Hexagonal Architecture (Ports & Adapters) adapted for Go idioms, ensuring that the business logic remains fully agnostic to external frameworks and message brokers.

• cmd/: Application entry points and dependency injection (Composition Root).
• internal/domain/: Core business models (Structs) and interfaces (Ports).
• internal/application/: Business use cases orchestrating the translation flow.
• internal/infrastructure/: Secondary adapters (Kafka Consumers/Producers, GCP Translation client).

🛠️ Technology Stack

• Core: Go 1.26+
• Messaging: Apache Kafka (github.com/segmentio/kafka-go)
• Cloud: Google Cloud Translation API (cloud.google.com/go/translate)
• Methodology: Strict TDD (Test-Driven Development) using the native Go testing package.

🚀 Development

To run the unit tests (following the Red-Green-Refactor TDD loop):

go test ./... -v

To compile the standalone binary:

go build -o bin/translator cmd/translator/main.go

---

👨‍💻 Author

Roberto Díaz
Senior Software Engineer & Architect

📄 License

Copyright (C) 2026 Roberto Díaz. All rights reserved.
Licensed under the GNU General Public License v3.0. See LICENSE for details.