Skip to content

juandelaf1/SkyCast

Repository files navigation

SkyCast

Enterprise climate intelligence platform integrating 5 data sources with periodic monitoring and alerting.

Full Case Study

Overview

SkyCast ingests data from 5 sources (AEMET, OpenWeatherMap, AirNOW, Open-Meteo, custom IoT sensors) into a unified PostgreSQL schema. A FastAPI backend processes and validates data, with Streamlit dashboards for interactive visualization.

Key Results

  • 158 automated tests across ingestion, validation, and API layers
  • Periodic monitoring (not real-time) via scheduled ETL
  • 5 data sources integrated under a unified schema
  • CI/CD pipeline with automated testing

Stack

FastAPI - PostgreSQL - Streamlit - Docker - Python - AEMET API

Architecture

Data Sources - Ingestion - PostgreSQL - FastAPI - Streamlit Dashboard

Limitations

  • Periodic monitoring rather than real-time streaming
  • IoT sensor integration experimental (simulated test data)
  • Single-server deployment

Detailed architecture decisions and trade-offs at juandelaf1.github.io/projects/sky-cast

About

Enterprise Climate Intelligence Platform - FastAPI, Streamlit, PostgreSQL, AEMET OpenData

Topics

Resources

Stars

1 star

Watchers

0 watching

Forks

Packages

 
 
 

Contributors