Download a dataset from a source like Kaggle, then clean and preprocess it. Identify missing or inconsistent data, decide how to handle these instances, and standardize/normalize data values.

The Data Cleaning Challenge

Design a SQL database schema for a hypothetical application. Implement this schema in a SQL database, populate the database with sample data, then write SQL queries to extract useful information.

The SQL Challenge

Using a dataset of your choice, create meaningful data visualizations (like line charts, bar charts, scatter plots, etc.) using a tool such as Matplotlib, Seaborn, or Plotly. Try to tell a story or reveal insights through your visualizations.

The Data Visualization Challenge

Build an ETL (Extract, Transform, Load) pipeline using a tool like Python's Pandas or a platform like Apache Beam. This could involve extracting data from a source, transforming the data to a useful format, and loading it into a database.

The ETL Pipeline Challenge

Train a machine learning model using a tool such as Scikit-learn or TensorFlow. Start with something simple like linear regression, then move on to more complex models. Evaluate the model's performance and try to improve it.

The Machine Learning Challenge

Given a dataset, formulate a hypothesis about the data, then use statistical methods to test this hypothesis. This could involve t-tests, chi-squared tests, or ANOVA, depending on the dataset and hypothesis.

The Hypothesis Testing Challenge

Design an A/B test for a hypothetical website or application. This would involve defining the goal of the test, deciding what to test, and determining how to measure the results.

The A/B Testing Challenge

Obtain a time series dataset and perform an analysis. This could involve identifying trends, seasonality, and other components of the time series. Forecast future points in the time series using methods like ARIMA or exponential smoothing.

The Time Series Analysis Challenge

Work with a larger dataset that cannot fit into memory. Use tools like Dask or Spark to perform computations on this dataset. This could involve cleaning the data, aggregating it, or applying machine learning algorithms.

The Big Data Challenge

Design a system to process real-time data, like tweets or IoT sensor readings. This could involve setting up a data stream, using a tool like Kafka or RabbitMQ, and processing the incoming data.

The Real-time Data Processing Challenge

A dashboard to track and visualize sales performance.

Sales Performance Dashboard

Analyze customer behavior to draw insights for business strategy.

Customer Behavior Analysis

Analyze market data to inform business decisions.

Market Research

Analyze healthcare data to find trends and patterns.

Healthcare Data Analysis

Conduct an A/B test to inform business decisions.

Data Analyst

Learning Resources

Interview Prep

Core Concepts

Code Challenges

Projects ideas

Best Practices