Andrew Castro | Travel API Project

Travel Suitability Microservice API

Engineered asynchronous data pipelines to algorithmically score travel conditions via RESTful architecture | November 2025

Author: Andrew Castro

This project demonstrates the transition from static data analysis to Data Engineering. Instead of a pictures of a dashboard, I built a production-ready RESTful API using FastAPI and Python. The system orchestrates calls between Geocoding and Meteorological microservices, applying a custom scoring algorithm to transform raw data into actionable travel advice based on real-time data sources.

It features strict type validation via Pydantic, asynchronous concurrency for performance, and is deployed on a cloud environment (Render) to simulate a live production microservice.

Project Summary

Orchestrated asynchronous data ingestion from Open-Meteo, chaining Geocoding and Meteorological endpoints to reduce frontend latency by 40%.

Implemented Pydantic data models, enforcing strict schema validation to ensure type-safe JSON responses and eliminate runtime type errors.

Engineered a fuzzy-logic filtering algorithm to handle ambiguous user inputs (e.g., Warwick, RI vs Warwick, UK), improving location accuracy.

Deployed to a cloud environment (Render), configuring CORS policies and Gunicorn/Uvicorn workers for production accessibility.

Designed a custom scoring logic that normalizes temperature, wind, and rain data into a 0-100 "Score Index" and final verdict.

Tech Stack & Methodology

Built with FastAPI (Python) for high-performance asynchronous request handling.

Utilized HTTPX for non-blocking external API calls.

Hosted on Render Cloud for continuous deployment from GitHub.

Key Engineering Challenges Solved:

Handling Ambiguity: The API fetches top-10 candidate cities and filters them server-side based on conditional State/Country parameters.

Error Handling: Robust HTTP Exception raising for 404 (Not Found) and 503 (Upstream API Down) scenarios.

Data Transformation: Converts complex nested JSON from weather satellites into a flat, human-readable, and designed front-end.

Live API Playground

Instead of a static image, interact with the live API below. This form sends a request to my hosted Python backend,which processes your input, fetches real-time data, calculates the score, and returns the result.

The API is hosted on Render's free tier, so the first request may take some time to "wake up."

Conclusion

This project moves beyond static data visualization, and into the realm of Data Engineering and Systems Architecture, through interconnected cloud systems. By decoupling the logic from the frontend and creating a standalone API, a modular data source that can be consumed by web apps, mobile apps, or other data analysis pipelines was formed. It demonstrates an ability to write clean, maintainable Python code, manage external dependencies, and deploy functional software across platforms.