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: Why vector search needs a hybrid approach (and how to build it). Read now
Dataquest Project Lab: A recap of 2025’s best projects—from Sassy Chatbots to IPO predictions. Watch recordings
From the Community: An end-to-end Nasdaq API projects, IoT sensor discussions, and productivity tools. Join the discussion
What We’re Reading: Why LLMs are confidently wrong and the coding concepts devs secretly struggle with. Learn more
Vector Search has a blind spot.
Semantic search is great at understanding meaning, but it struggles with specifics—like filtering by date or finding rare keywords. In this tutorial, we upgrade your vector database skills by adding Metadata Filtering and Hybrid Search.
You’ll learn how to:
-
Filter effectively: Constrain searches by date or category in ChromaDB.
-
Combine strategies: Blend vector scores with traditional keyword search.
-
Challenge assumptions: See our surprising data on when hybrid search actually adds value.
Dataquest Project Lab: The 2025 Rewind
This year, we went beyond theory and shipped real code. From Sassy Chatbots and Streamlit apps to predicting IPO gains with TensorFlow and analyzing Star Wars data, we covered the full stack.
Missed a session? Catch up on all 20+ guided projects in our archive. Watch the replays here.
From the Community
Exploring Financial Data Using Nasdaq Data Link API: Tomaz’s end-to-end project efficiently covers all the steps of a professional data analysis workflow—from retrieving data via API to conducting thorough data exploration, cleaning, and insight extraction.
Community No-Lose Lottery Winner: Discover who won the latest Community No-Lose Lottery and learn how you can participate in its next round for a chance to win.
Online Notepad for Keeping Your Work Organized: Alihan shared a free, user-friendly online editor designed to help you track data object transformations and take helpful notes throughout your data science projects.
Using IoT Sensor Data for Data Science Projects: Aria is looking to use real-world IoT sensor data as a dataset for data science learning and practice and asks Community members to share their experience and discuss potential challenges.
What We're Reading
Seven Things Developers Don’t Really Understand: A fun and humbling read on the parts of coding most of us quietly struggle with — from boolean logic and threading to Unicode and time zones. Even experienced engineers will recognize a few blind spots here.
Why LLMs Can Sound Confident and Be Wrong: MIT researchers uncovered a subtle flaw in how LLMs learn patterns. Instead of reasoning, models can latch onto familiar sentence structures and return convincing but incorrect answers.
How People Actually Use AI Agents: A new study from Perplexity and Harvard shows AI agents are used far more for deep cognitive work and research than for everyday task automation. An interesting counterpoint to the usual “AI will book your flights” narrative.
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.