skill~150 mindifficulty

Vector Databases

To search over many documents fast, you need a purpose-built store for vectors.

1

Learn it

Work through the Chroma introduction to spin up a local, open-source vector store in a few lines: create a collection, add documents (embeddings handled for you), and query for nearest neighbors. Explore the repo for metadata filtering and persistence.

By the end you’ll be able to

  • Store embeddings with metadata and query by similarity
  • Filter results by metadata alongside vector search

Free resources

2

Practice

Load a folder of text notes into a Chroma collection with metadata (source, date), then run similarity queries and filter to a single source.

3

Build it — your output

A local vector store populated with your own documents and a query script.

4

Prove it — check yourself

You’re ready to move on when you can honestly say yes to each:

  • Can you add documents to a collection and query the nearest matches?
  • Can you combine a similarity search with a metadata filter?
  • Can you explain what an approximate nearest neighbor index buys you?

Do it with AI

Use a free AI tool as your tutor and pair-programmer. Copy a prompt below into Claude, ChatGPT, or Gemini (for building, try Cursor or Copilot).

The habit: Try it yourself first — a little struggle is where the learning actually happens. Then ask for a hint before the answer — never copy blindly.

  • Understand itClaude / ChatGPT / Gemini

    I'm a beginner learning "Vector Databases". Explain it in simple terms with a real-world analogy, then ask me ONE question to check my understanding — don't reveal the answer.

  • PracticeClaude / ChatGPT / Gemini

    Give me 3 short practice problems on "Vector Databases" (my goal: store embeddings with metadata and query by similarity), from easy to hard. Don't include the solutions yet — I'll try them first, then ask you to check.

  • Get unstuckClaude / ChatGPT / Gemini

    I'm stuck on this exercise about "Vector Databases": [paste your attempt here]. Give me ONE small hint to move forward — do NOT give me the full solution.

  • Build & reviewCursor / Copilot

    Here is my work for a "Vector Databases" task: [paste your code or answer]. Review it: point out what's wrong and WHY, and suggest improvements — but let me make the fixes myself.

  • Prove itClaude / ChatGPT / Gemini

    Quiz me on "Vector Databases". Ask me 5 questions one at a time, wait for my answer each time, then tell me if I'm right and briefly why.

My notes