The Dataquest Download
Level up your data and AI skills, one newsletter at a time.
Hello, Dataquesters!
Here’s what we have for you in this edition:
Top Read: Build a multi-provider LLM gateway in Python and learn the patterns that make AI apps easier to maintain, scale, and extend. Learn more
From the Community: Statistical analysis in the Winning Jeopardy project, ideas to make the Food Ordering App more realistic, tips for stronger project storytelling, and why Python and SQL work better together. Join the discussion
What We’re Reading: How AI is reshaping business intelligence, why machine unlearning remains one of AI’s biggest trust challenges, and what Microsoft’s Work Trend Index 2026 reveals about the real skills that matter as AI becomes part of everyday work. Learn more
Top Read
What happens when you need to switch AI providers? If your application is tightly coupled to a single API, even a simple change can require significant rework. That’s why many AI teams use a gateway layer between their application and LLM providers.
In this hands-on project, you’ll build a multi-provider LLM gateway from scratch using Python. You’ll create a single interface that works across providers like Anthropic and TogetherAI, handle authentication and response differences behind the scenes, and learn the architectural patterns that make AI applications easier to maintain, scale, and extend.
From the Community
Winning Jeopardy: To identify the most frequent topics on the show, Daniel formulated a hypothesis and tested it using a chi-squared test, then went a step further by conducting a thorough bigram analysis to provide a more focused view of topic frequency. An excellent example of diving deep into the data to produce robust and meaningful results.
Building a Food Ordering App: Alla suggests implementing features such as discount codes, order history, and the ability to save orders to a file to make the app even more realistic, as well as enhancing the checkout process by asking users to confirm their orders before finalizing them.
Data Science Project Storytelling: Linky emphasizes the importance of maintaining a relevant narrative throughout a project by clearly stating its purpose, providing necessary context and background, outlining the methodology, explaining the thought process, discussing observations, and drawing final findings.
Exploring Community-Generated Data: Sarah offers the insightful suggestion of analyzing data science-focused, community-generated datasets, such as those from Stack Exchange, as a practical way to understand real-world trends in data science and potentially build an outstanding portfolio project.
Python vs SQL: Mamta highlights that using Python and SQL together makes data work much more powerful and effective, with SQL excelling at extracting and organizing data while Python enables deeper analysis and the creation of compelling visualizations.
What We're Reading
BI Is Dead. Long Live BI: Dashboards aren’t disappearing, but the way people interact with data is evolving. This article explores how AI is reshaping business intelligence and what the next generation of BI tools may look like.
A New Framework for Auditing Machine Unlearning: Researchers found that many popular machine unlearning techniques may not fully remove the data they’re supposed to forget. A fascinating look at one of the biggest trust and privacy challenges in AI.
Microsoft Work Trend Index 2026: This report suggests that getting value from AI depends less on prompt tricks and more on how people redesign workflows, collaborate with AI agents, and review outputs. A useful look at the skills that may matter most as AI becomes part of everyday work.
Give 20%, Get $20: Time to Refer a Friend!
Give 20% Get $20
Now is the perfect time to share Dataquest with a friend. Gift a 20% discount, and for every friend who subscribes, earn a $20 bonus. Use your bonuses for digital gift cards, prepaid cards, or donate to charity. Your choice! Click here
High-fives from Vik, Celeste, Anna P, Anna S, Anishta, Bruno, Elena, Mike, Daniel, and Brayan.