Everyone’s doing it! AI tools have been around for quite a while, but they are definitely having a moment, and people are enjoying the thrill of discovery with these new tools.
I thought it would be fun to show a quick and easy way you can use Google Gemini to create a simple Python app that tells you the weather.
Enter This
Create a Python script that connects to the National Weather Service API and gives the next week’s forecast for a specified location.
Result
import requests
import json
# Define user-agent header (NWS API requires a custom User-Agent identifying your app)
# Format: (your-app-name, your-contact-email@example.com)
HEADERS = {
"User-Agent": "(MyWeatherApp/1.0, contact@example.com)",
"Accept": "application/geo+json"
}
def get_nws_forecast(latitude: float, longitude: float):
"""
Retrieves the 7-day weather forecast for a given latitude and longitude
using the National Weather Service API.
"""
# Step 1: Query the /points endpoint to get local grid office metadata
points_url = f"https://api.weather.gov/points/{latitude:.4f},{longitude:.4f}"
try:
response = requests.get(points_url, headers=HEADERS, timeout=10)
response.raise_for_status()
point_data = response.json()
# Extract the specific forecast URL endpoint and location details
forecast_url = point_data["properties"]["forecast"]
location_info = point_data["properties"]["relativeLocation"]["properties"]
city = location_info.get("city")
state = location_info.get("state")
print(f"\nFetching forecast for: {city}, {state} ({latitude}, {longitude})\n" + "=" * 60)
# Step 2: Fetch the detailed forecast from the endpoint grid URL
forecast_response = requests.get(forecast_url, headers=HEADERS, timeout=10)
forecast_response.raise_for_status()
forecast_data = forecast_response.json()
# Extract forecast periods (contains day/night detailed breakdowns)
periods = forecast_data["properties"]["periods"]
for period in periods:
name = period["name"]
temp = period["temperature"]
unit = period["temperatureUnit"]
short_forecast = period["shortForecast"]
detailed_forecast = period["detailedForecast"]
wind_speed = period["windSpeed"]
wind_dir = period["windDirection"]
print(f"--- {name} ---")
print(f"Temperature : {temp}°{unit}")
print(f"Wind : {wind_speed} {wind_dir}")
print(f"Overview : {short_forecast}")
print(f"Details : {detailed_forecast}\n")
except requests.exceptions.HTTPError as http_err:
print(f"HTTP error occurred: {http_err}")
except requests.exceptions.RequestException as req_err:
print(f"Request failed: {req_err}")
except KeyError as key_err:
print(f"Error parsing response JSON key: {key_err}")
if __name__ == "__main__":
# Example: New York City Coordinates (40.7128° N, -74.0060° W)
# Note: NWS API only covers US regions/territories.
lat = 40.7128
lon = -74.0060
get_nws_forecast(lat, lon)
Final Details
Of course you also have to know how to install requests and run the Python script on your computer! But Gemini did the coding for you. Sweet!