Skip to main content
grasp.study
About
Explore
Pricing
Log in
Try it for free
Back to course
👋
Welcome to a Grasp lesson
Create a free account
to keep studying and create your own study path.
Elite ML Engineer from Day 1
·
Module 10
Retrieval-Augmented Generation Systems
1
RAG Architecture for Mitigating LLM Limitations
Explain the RAG architecture and how it mitigates LLM limitations like knowledge cutoffs and hallucinations.
2
Mastering Document Chunking Strategies
Implement various document chunking strategies, considering trade-offs between chunk size and overlap.
3
Sentence-Transformers for Text Embedding
Generate high-quality embeddings for text chunks using sentence-transformer models.
4
Vector Database Setup for Document Embeddings
Set up a vector database (e.g., FAISS, ChromaDB) to store and index document embeddings.
5
Semantic Search with Cosine Similarity
Implement semantic search using cosine similarity to retrieve relevant context for a given query.
6
Building a RAG Pipeline with LangChain/LlamaIndex
Build a complete RAG pipeline connecting retrieval and generation using a framework like LangChain or LlamaIndex.
7
Hybrid Search: Combining Sparse and Dense Retrieval
Explain sparse retrieval methods (e.g., BM25) and implement a hybrid search approach combining dense and sparse retrievers.
8
RAG System Evaluation: Retrieval and Generation Metrics
Evaluate a RAG system using retrieval metrics (e.g., hit rate, MRR) and end-to-end generation quality.
Previous module
Adapting LLMs: Fine-Tuning and Alignment
Next module
Agentic AI Systems and Orchestration