---
title: "Discussion on \"Exploring GenAI Retrieval with a Resume: Keyword, Semantic, and Hybrid Search\" | Hashnode"
description: "Discussion on \"Exploring GenAI Retrieval with a Resume: Keyword, Semantic, and Hybrid Search\". A GenAI learning project exploring resume retrieval using embeddings, keyword search, semantic search, and hybrid search with document chunking."
url: "https://hashnode.com/posts/exploring-genai-retrieval-with-a-resume-keyword-semantic-and-hybrid-search/6ac2492734776c6712e86346"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Exploring GenAI Retrieval with a Resume: Keyword, Semantic, and Hybrid Search](https://dev-praveen.hashnode.dev/exploring-genai-retrieval-with-a-resume-keyword-semantic-and-hybrid-search?utm_source=hashnode&utm_medium=feed)

A GenAI learning project exploring resume retrieval using embeddings, keyword search, semantic search, and hybrid search with document chunking.

Oct 4 · 6 min read · dev-praveen.hashnode.dev  
Read the article

-   [#generative-ai](https://hashnode.com/tag/generative-ai)
-   [#genai](https://hashnode.com/tag/genai)
-   [#rag](https://hashnode.com/tag/rag)
-   [#hybrid-search](https://hashnode.com/tag/hybrid-search)
-   [#semantic-search](https://hashnode.com/tag/semantic-search)
-   [#keyword-search](https://hashnode.com/tag/keyword-search)
-   [#embeddings](https://hashnode.com/tag/embeddings)
-   [#document-search](https://hashnode.com/tag/document-search)

## Written by

[praveenkulharee](https://hashnode.com/@praveen-dev)

@praveen-dev · Delhi, India

  
[View profile](https://hashnode.com/@praveen-dev)

## Responses 2

Markdown supported · ⌘↵ to post

[praveenkulharee](https://hashnode.com/@praveen-dev)

[Oct 5](https://hashnode.com/posts/exploring-genai-retrieval-with-a-resume-keyword-semantic-and-hybrid-search/6ac2492734776c6712e86346/comment/6ac368b694c6c9b912d4011e)

Thanks for the valuable feedback! The 0.7/0.3 blend is part of my learning implementation, so I’ll test query variations and compare raw, normalized, and rank-fusion approaches next.

[Ahmet Özel](https://hashnode.com/@ahmetozel)

[Oct 4](https://hashnode.com/posts/exploring-genai-retrieval-with-a-resume-keyword-semantic-and-hybrid-search/6ac2492734776c6712e86346/comment/6ac2918225455dedc00c7ccf)

The 0.7 semantic / 0.3 keyword blend is a useful experiment, but the scales determine what those weights actually mean. Your keyword score counts matching query tokens, while cosine similarity lives on a different scale; a longer query can therefore increase the keyword contribution without any change in relevance.

I would print both raw components beside the combined score and try the same question with a few unnecessary words added. Normalizing the keyword component or comparing rank fusion would help separate a deliberate weighting choice from an accidental consequence of query length.
