skills.rest
Jhabbig
JhabbigCOMMUNITY
Explore all skills by this author →
GitHub

embedding-strategies

Select embedding models and chunking strategies for semantic retrieval pipelines.

Updated Apr 5, 2026
Search Tags:#semantic search#embeddings#rag#vector search#multilingual#chunking#retrieval evaluation
One-click install
npx skills add https://github.com/Jhabbig/Habbig --skill embedding-strategies-jhabbig
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: embedding-strategies
Source: https://github.com/Jhabbig/Habbig/tree/main/.claude/plugins/wshobson/llm-application-dev/skills/embedding-strategies
Command: npx skills add https://github.com/Jhabbig/Habbig --skill embedding-strategies-jhabbig

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you choose, tune, and evaluate embeddings so semantic search and RAG systems retrieve the right information instead of noisy matches.

Core Features & Use Cases

  • Model selection across general, code, finance, legal, and multilingual embedding models.
  • Chunking strategies for tokens, sentences, markdown sections, and source code.
  • Retrieval quality evaluation with precision, recall, MRR, and NDCG for comparing pipelines.
  • Use it when building vector search for chatbots, document search, code search, or domain-specific knowledge bases.

Quick Start

Use the embedding-strategies skill to recommend the best embedding model, chunking plan, and evaluation approach for your corpus.

Frequently Asked Questions about embedding-strategies

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose the best embedding model for RAG?▼

To choose the best embedding model for RAG, compare general, code, finance, legal, and multilingual models against your specific corpus. Evaluate retrieval quality using precision, recall, MRR, and NDCG to identify the most effective option for semantic search.

What chunking strategy should I use for semantic search?▼

Chunking strategies for semantic search depend on your data type. Apply token-aware chunking for general text, sentence or markdown section splitting for documents, and specialized source code chunking to preserve context and improve vector search retrieval accuracy.

How do I evaluate retrieval quality in a RAG pipeline?▼

Evaluate retrieval quality in a RAG pipeline by measuring precision, recall, MRR, and NDCG. Comparing these metrics across different embedding models and chunking strategies helps identify which pipeline configuration retrieves the most relevant information with minimal noise.

Can I use specialized embeddings for domain-specific document search?▼

Yes, you can use specialized embeddings for domain-specific document search. Select models tailored for code, finance, or legal corpora to improve semantic retrieval. Apply preprocessing, normalization, and query-document prefix handling to optimize domain-specific vector search results.

Why does my vector search return noisy matches?▼

Vector search returns noisy matches when embedding models and chunking strategies are poorly aligned with your corpus. Fix this by applying token-aware chunking, proper preprocessing, query-document prefix handling, and evaluating retrieval quality with NDCG and MRR metrics.

Do I need multilingual embedding models for non-English corpora?▼

Yes, you need multilingual embedding models for non-English corpora to maintain semantic search accuracy. Selecting a multilingual model ensures consistent vector search retrieval quality across different languages within your RAG pipeline or document search workflow.

Related Skills

Explore More Software Engineering Skills→
lucifertrjlucifertrj
17

nltk-nlp-text-cleaning

Clean, tokenize, and normalize English text with NLTK for NLP pipelines.

Community
Advanced
dandudzidandudzi
2

rag-implementation

Design and implement RAG pipelines with chunking, embeddings, and vector stores.

Community
Advanced
anukkrit149anukkrit149

rag-implementation

Build Retrieval-Augmented Generation systems grounding LLM answers in vector databases.

Community
Advanced
JSai23JSai23
1

vault-sem-search-a

Enable semantic search of local vault notes using qmd with BM25, vector search, and LLM re-ranking.

Community
Intermediate