Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

Scaling up RAG with Vector Databases using JavaScript

Go to class Write review

Overview

Build and query a vector database for RAG: preprocess documents, store chunk embeddings in ChromaDB, retrieve relevant chunks with metadata filters and weighting, craft multi-chunk prompts, manage collection updates, and scale ingestion with batch strategies.

Syllabus

Unit 1: Scaling Up RAG with Vector Databases: Document Chunking in JavaScript

Implementing Word-Based Text Chunking
Refining Text Chunking with Sentence Boundaries in JavaScript
Chunking Text with Overlap in JavaScript
Organizing Text Chunks with Metadata in JavaScript
Enhancing Text Chunking with Keyword Detection

Unit 2: Storing and Managing Text Chunks in Vector Databases with JavaScript

Exploring Vector Databases with ChromaDB in JavaScript
Building a ChromaDB Collection with JavaScript
Dynamically Managing ChromaDB Collections in JavaScript
Efficiently Managing ChromaDB Collections with Keyword-Based Deletion

Unit 3: Retrieving Relevant Chunks and Building LLM Prompts in JavaScript

Enhance the retrieveTopChunks Function for Semantic Similarity Retrieval
Integrating Contextual Prompts with LLMs in JavaScript
Implementing Metadata Filtering for RAG Vector Retrieval
Incorporating Distance Threshold in Retrieval Process

Unit 4: Metadata-Based Filtering in RAG Systems with JavaScript

Implement Metadata Filtering in JavaScript
Metadata Enhanced Search in JavaScript
Enhancing Document Search with Category and Date Filters
Implementing a Fallback Mechanism for Metadata-Based Search in JavaScript

Reviews

Start your review of Scaling up RAG with Vector Databases using JavaScript