The Dataquest Download

Build data and AI skills – one newsletter at a time

Our Hello Summer Lifetime deal closes soon. Get 57% off Lifetime and unlock every course, path, and project with one payment, so you can keep building skills all summer and beyond. Save now

Here’s what’s inside:

• Top Read: Build real data cleaning skills with a hands-on pandas project based on 50,000 used car listings. Practice handling missing values, inconsistent formats, outliers, and messy real-world data. Learn more

• Webinar Recording: Build Word Raider, an interactive Python word game, and practice reading files, validating input, and writing game logic with loops and conditionals. Watch now

• 
  

  Product Update: We’ve added a new Deploying to Cloud course to the Data Engineer path. Alongside our AWS content, you can now get hands-on practice deploying data projects to Google Cloud through guided walkthroughs. View course

• From the Community: A strong movie ratings analysis project, practical advice on what makes data science projects stand out, and tips for writing clearer, more helpful app error messages. Join the discussion

• What We’re Reading: Why smaller AI workflows can beat giant prompts, how average ML metrics can hide real problems, new uses for Direct Preference Optimization, and why judgment matters more in an AI-powered workplace. Learn more

Top Read

The data cleaning skills every data analyst needs (hands-on project)

Data cleaning is where much of a data analyst’s real work happens. Before you can build dashboards, create models, or uncover insights, you need to deal with missing values, inconsistent formats, outliers, and messy datasets.

In this hands-on project, you’ll work with 50,000 real-world used car listings and practice the same data cleaning techniques analysts use every day. You’ll clean and transform raw data with pandas, identify quality issues, and uncover pricing trends hidden beneath the noise. If you want practical experience with one of the most important skills in analytics, start here.

Webinar Recording

Missed our last Project Lab? The recording is now available.

In this session, you’ll learn how to build “Word Raider,” an interactive word-guessing game using Python. You’ll structure a complete application from scratch, read external files, validate user input, and implement game logic using loops and conditionals.

If you want hands-on practice turning core Python concepts into a fully working project, this walkthrough is a great place to start.

Product Update

We’ve added a new Deploying to Cloud course to the Data Engineer path, expanding on our existing cloud deployment coverage.

Alongside our existing AWS deployment material, this new course provides hands-on practice deploying data projects to Google Cloud Platform through guided walkthrough projects. It’s designed to help build practical cloud deployment skills across more than one major cloud platform.

From the Community

• Analyzing Movie Ratings: Daniel’s project demonstrates an efficiently organized structure, an easy-to-follow data narrative, well-documented code, insightful data visualizations, and detailed conclusions that reveal correlations between IMDb movie ratings and vote counts.

• Key Components of Standout Data Science Projects: Linky highlights several essential elements of exceptional data science projects, including a clear structure, thorough data cleaning to improve data quality, curiosity-driven analytical depth, a well-presented and logical workflow, and effectively communicated findings.

• Improving Error Notifications in Apps: Alla outlines best practices for designing actionable error notifications when developing apps, such as incorporating relevant user inputs in error messages and clearly explaining the specific cause of the error to help users resolve issues more efficiently.

What We're Reading

• Stop Using LLMs Like Giant Problem Solvers: Instead of throwing entire problems at an AI model, break them into smaller steps. This article explains why structured workflows often outperform bigger prompts.

• Why Average ML Metrics Can Be Misleading (MIT): A model that performs best overall can still fail badly for specific groups or environments. MIT researchers explain why looking beyond aggregate metrics is critical for real-world machine learning.

• Direct Preference Optimization Beyond Chatbots: This article explores how a document OCR model used failed outputs as training examples, showing how techniques like Direct Preference Optimization can improve models beyond chatbot applications.

• Microsoft Work Trend Index 2026: As AI agents take on more execution work, human judgment is becoming more important. Microsoft’s latest report highlights quality control and critical thinking as two of the most valuable skills in an AI-powered workplace

Learning with Friends

Bring an accountability partner along on your learning journey. Refer a friend, and they’ll enjoy an extra 20% off when they subscribe, while you earn $20. It’s a win-win for everyone. Learn together, stay motivated, and use your bonuses for digital gift cards, prepaid cards, or charity donations!