This is a submission for the Xano AI-Powered Backend Challenge: Production-Ready Public API
What I Built
I built Astryx Documents API, a production-ready backend designed to ingest aeronautics documents, structure them, split them into semantic chunks, and store vector embeddings for future AI-powered search and retrieval.
The API is designed for real-world use cases such as technical documentation analysis, maintenance manuals, and knowledge retrieval systems.
It is fully built with Xano, secured with token-based authentication, and structured to be scalable and maintainable.
API Documentation
The API exposes secure, private endpoints that can be consumed by third-party applications.
Core Endpoints
-
POST
/documents- Adds a document record with metadata.
-
POST
/add_document_with_vectors- Ingests a document
- Splits its content into chunks
- Stores vector embeddings linked to each chunk
Security
- Token-based authentication (Bearer token required)
- Private API group configuration
This makes the API ready for controlled public or partner access.
Demo
📹 Demo video & screenshots:
👉 Google Drive link: https://drive.google.com/drive/folders/1hem6CqRIFdi62VNkmUicp1Gwa_lygJNl?usp=sharing
Github :https://github.com/AsamaeS/Astryx
The demo shows:
- API execution in Xano
- Document insertion
- Automatic chunk creation
- Vector storage in the database
The AI Prompt I Used
I used AI to generate an initial backend structure and XanoScript endpoints for document ingestion and storage.
The goal of the prompt was to accelerate backend creation and then manually refine it to meet production standards.
How I Refined the AI-Generated Code
The AI-generated code required significant refinement to become production-ready:
- Fixed invalid XanoScript syntax that did not compile
- Refactored endpoints to respect Xano’s execution model
- Removed unsupported control structures
- Added secure authentication requirements
- Designed a clear document → chunk → vector data pipeline
- Improved maintainability and clarity of the API logic
This process highlighted how human expertise is essential to transform AI-generated code into reliable production systems.
My Experience with Xano
Xano made it possible to move extremely fast from idea to production-ready backend without managing infrastructure.
The biggest challenge was mastering strict XanoScript syntax, but once understood, it enabled clean, scalable backend logic with excellent performance and security.
Xano proved to be a powerful platform for building real-world APIs enhanced by AI workflows.