đź“‹ Lab Reference

Associated Lab File: vector-search-1.lab.js

🚀 Goal: Semantic Search That Wows

Your business wants to make searching for the perfect stay effortless and intelligent. Imagine a guest typing a natural-language query and instantly seeing smart, intent-aware suggestions—helping them find their dream destination even if they don’t use the exact words. As the backend engineer, you’re about to bring this next-level search to life with MongoDB Atlas Vector Search.

Harness the power of MongoDB Atlas Vector Search to build a semantic search feature your users will love!

🧩 Exercise: Semantic Search Like a Pro

  1. Open the File
    Head to server/src/lab/ and open vector-search-1.lab.js.

  2. Find the Function
    Locate the vectorSearch function.

  3. Shape the Pipeline

    • Add a $vectorSearch stage using your vector index.
    • Use the description field as the vector search path.
    • Pass the user’s query string as the query parameter.
    • Add a filter on property_type for more relevant results.
    • Set numCandidates to 100 and limit to 10 in the $vectorSearch stage.

🚦 Test Your API

  1. Go to server/src/lab/rest-lab.
  2. Open vector-search-1-lab.http.
  3. Click Send Request to call the API.
  4. Confirm the response contains the expected, semantically relevant results.

🖥️ Frontend Validation

Type a natural-language query (e.g., "best view in hawaii") in the search bar and watch smart, relevant suggestions appear—powered by AI and vector search!

Check Exercise Status:
Go to the app and check if the exercise toggle shows green, indicating your implementation is correct.

With this step, you’re not just building a feature—you’re enabling a new era of discovery and delight for your users.
Ready to wow your guests with semantic search? Let’s get started!

vector-search-1-lab