Zypp Skill Test: Dutch Weather Analysis

Skilltest designed for Zypp's Recruitment process

Time Cap: 4 hours
Coding Skills: Python, Pandas, Requests
Technical Skills: API building, Data Modeling
Fundamental Skills: Data Integration, Data Analysis, Automation

Table of Contents

1. Introduction Skilltest
   • Our expectations
2. The test: Dutch Weather Analysis
   • Case introduction
   • Instructions
   • Guidelines
3. Part 1: Data Integration
   • Question 1: Create station measurement data
   • Question 2: Create weather station data
   • Question 3: Store data in an SQL database
   • Question 4: Create an ERD of the SQL database
4. Part 2: Data Analysis
   • Question 5: Find the weather station with the highest temperature
   • Question 6: Calculate the average temperature
   • Question 7: Identify the coldest time of the day
   • Question 8: Locate the weather station in the North Sea
   • Question 9: Create a bar chart of average temperature
5. Part 3: Automation
   • Question 10: Automate data collection

Introduction

Welcome to this Zypp's skill test! This test is designed to assess your proficiency in Python programming and problemsolving with a focus on data integration, data analysis, and automation.

You may have arrived here as you're heading into the second part of our recruitment process. In this second part, we're eager to explore the world of coding together. Our focus will be on observing your coding skills, understanding your thought process, and discovering how you solve problems. In essence, we want to initiate an engaging dialogue about code. This skill test provides a unique opportunity for you to showcase your coding abilities. This repository will serve as an ideal starting point for you to demonstrate your coding skills, but also for us to show what kind of projects you will encounter during your career at Zypp.

Our expectations

This project is not designed to be a pass or fail test. There are multiple ways to solve the questions, we want to see how you solve them. Based on your answers, we will:

• determine your 'base' level of skills.
• have a follow up conversations, were we will talk about the project and the choices you have made.

---

The test: Dutch Weather Analysis

Case introduction

The Buienradar API provides data from all weather stations in the Netherlands,updated 3 times an hour. You can query the data via the endpoints json or xml, and it will return the current weather data for each station. The goal is to answer some data analysis questions, but we need data for a full day from buienradar. In order to answer the questions, the candidate needs to be able to:

1. create a script for importing the data and exporting it to a database.
2. model the data in the database, by relating the tables and type casting the data in the proper format.
3. automate the script to collect 1 day of data from the buienradar API.

The test is decomposed into 3 sections.

Instructions:

1. This repository contains a set of Python coding tasks related to data integration, data analysis, and automation.
2. Fork this repository to your GitHub account to get started with the test.
3. Complete the tasks within the specified time frame, there is a 4 hours timecap.
4. Commit your solutions to your forked repository.
5. In the test you will create a database, please be sure to also commit the database to the repository.
6. Send the link to your forked repository to hello@zypp.io

Guidelines:

• You are encouraged to use Python 3.10 and any relevant libraries or frameworks for the tasks.
• Each section starts with a time estimation. This is not a hard requirement, but meant to give you clarity in how much time a topic should take.
• Feel free to reach out to the recruitment team if you have any questions or need clarification.
• The repository is public, and other candidates may also have access to it. Avoid sharing your solutions or collaborating with others during the test to maintain fairness.

Best of luck with the skill test! We are excited to see your engineering skills in action.

---

Part 1: Data Integration

Time Estimation: 2 hours
In this section, you will design the datamodel and ETL code for answering the questions in Part 2. This step is the most time consuming in this project, or general data-analysis projects.

Question 1: Create a dataset with the following information about the weather station measurements.

• measurementid (not in dataset by default)
• • timestamp
• temperature
• groundtemperature
• feeltemperature
• windgusts
• windspeedBft
• humidity
• precipitation
• sunpower
• stationid

Question 2: Create a dataset with the information about the weather stations:

• stationid
• stationname
• lat
• lon
• regio

Question 3: Store the measurements data and the station data in an SQL database. Use .sqlite for the database. Consider using index, Primary Key, and defining the relationship between the two tables.

Question 4: Create an ERD of the SQL database you created. Tip: you can use draw.io for making the diagram.

---

Part 2: Data Analysis

This section is about performing data analysis on your gathered data.

Time Estimation: 1 hours
In this part you are required to answer questions based on data collected in step 1.

Question 5: Which weather station recorded the highest temperature?

Question 6: What is the average temperature?

Question 7: What is the station with the biggest difference between feel temperature and the actual temperature?

Question 8: Which weather station is located in the North Sea?

Question 9: Create a bar chart, using a python package of your choice, with the temperature for each station.

---

Part 3: Automation

time estimation: 1 hour This section is about describing how you would automate this project (Theoretical). Feel free to use a flowchart, showing the steps of your approach.

Question 10: Describe how you would populate the database with all with all measurements for a specific day. In other words, the script you created in question 1 should be used to fetch the weather station data multiple times per hour, as the weather station data is updated every 20 minutes.