🚀 Introducing NeuraDB: A Lightweight In-Memory Vector Store for Fast Similarity Search in TypeScript
Hey devs! 👋
Today, I’m excited to open-source a project I’ve been building: NeuraDB, a zero-dependency, in-memory vector store optimized for blazing-fast similarity search—built entirely in TypeScript.
🔍 What is NeuraDB?
NeuraDB is a developer-first vector store designed for quick prototyping, experimentation, and real-world use in Node.js environments. It’s ideal for LLM apps, embedding-based search engines, RAG workflows, and more.
import { NeuraDB } from “neuradb”;
const db = new NeuraDB();
db.addDocument({
id: “1”,
content: “Hello world”,
embedding: [0.1, 0.2, 0.3],
metadata: { type: “greeting” }
});
const results = db.search([0.1, 0.2, 0.3], {
similarityMethod: “cosine”,
limit: 3
});
console.log(results);
✨ Features
•	🧠 Supports multiple similarity methods: cosine, euclidean, dot
•	⚡ Zero dependencies — small footprint, blazing fast
•	🏷️ Rich metadata filtering for selective queries
•	🧩 TypeScript native for a great DX
•	🧮 In-memory, lightweight alternative to complex vector databases
📦 Why build this?
I wanted something simple, fast, and local that just works. Most vector DBs out there are:
•	Heavily bloated with dependencies 🧱
•	Require setup (Docker, Postgres, etc.) 🐘
•	Overkill for quick LLM app experiments 🤯
NeuraDB is intentionally minimal but extensible.
🔭 Future Roadmap
Here’s what’s coming:
•	HNSW (approximate nearest neighbors)
•	Cold/hot memory tiering
•	SQL-like filtering on metadata
•	Edge-ready builds for mobile/browser
•	RAG utilities for LLM developers
If you have ideas or want to contribute, feel free to jump in!
📚 Learn More
GitHub: <a href="https://github.com/your-username/neuradb" target="_blank" rel="noopener nofollow">https://github.com/your-username/neuradb</a>
NPM: <a href="https://www.npmjs.com/package/neuradb" target="_blank" rel="noopener nofollow">https://www.npmjs.com/package/neuradb</a>
🫶 I’d love your feedback
If you’re building AI-native apps, embedding search, or anything vector-related—try NeuraDB out and let me know what you’d like to see next!
#vectorsearch #typescript #llm #opensource #neuradb #developers

The official daily.dev AI community. Led by thought leaders and AI experts. Stay tuned for insights, news and discussions. 

NeuraDB is a new open-source, zero-dependency vector store built in TypeScript for fast similarity search operations. It supports multiple similarity methods (cosine, euclidean, dot product), offers rich metadata filtering, and runs entirely in-memory. Designed as a lightweight alternative to complex vector databases, it's particularly suited for LLM applications, RAG workflows, and embedding-based search engines. The project aims to simplify vector operations for developers who need quick prototyping without the overhead of traditional vector database setups.

🚀 Introducing NeuraDB: A Lightweight In-Memory Vector Store for Fast Similarity Search in TypeScript

Looks great! But the repository is not accessible.

Uhh you forgot to replace the chatgpt placeholder github URl with you actual github URL.
Same for the npm package