Runs in your browser — no install. (Works once the repo is public.)

In [ ]:

# --- Colab setup (skipped when running locally) ---
import os, sys
if 'google.colab' in sys.modules and not os.path.exists('data/sample_corpus.json'):
    !git clone -q https://github.com/mdhabibi/llm-search-handbook.git
    %cd llm-search-handbook
    !pip -q install -r requirements.txt

# --- Colab setup (skipped when running locally) --- import os, sys if 'google.colab' in sys.modules and not os.path.exists('data/sample_corpus.json'): !git clone -q https://github.com/mdhabibi/llm-search-handbook.git %cd llm-search-handbook !pip -q install -r requirements.txt

Chapter 0 — 5-Minute Pipeline Tour¶

See the destination before learning each part. We stand up a search system in a few cells and ask it questions. Every piece here gets built from scratch in later chapters.

Setup¶

pip install sentence-transformers rank-bm25

In [ ]:

# Bootstrap: locate repo root and import shared helpers
import sys, os, json
d = os.getcwd()
while d != os.path.dirname(d) and not os.path.exists(os.path.join(d, 'data', 'sample_corpus.json')):
    d = os.path.dirname(d)
ROOT = d; sys.path.insert(0, os.path.join(ROOT, 'src'))
import numpy as np
from corpus import load_corpus

# Bootstrap: locate repo root and import shared helpers import sys, os, json d = os.getcwd() while d != os.path.dirname(d) and not os.path.exists(os.path.join(d, 'data', 'sample_corpus.json')): d = os.path.dirname(d) ROOT = d; sys.path.insert(0, os.path.join(ROOT, 'src')) import numpy as np from corpus import load_corpus

Build a search system in 3 lines¶

SearchStack bundles the whole course (chunk → hybrid retrieve → optional re-rank/generate). We'll use it search-only for the tour.

In [ ]:

from search_stack import SearchStack
from sentence_transformers import SentenceTransformer
embedder = SentenceTransformer('all-MiniLM-L6-v2')
stack = SearchStack(embedder.encode, use_hybrid=True).ingest(load_corpus())
print('Ready. Ask it anything about the corpus.')

from search_stack import SearchStack from sentence_transformers import SentenceTransformer embedder = SentenceTransformer('all-MiniLM-L6-v2') stack = SearchStack(embedder.encode, use_hybrid=True).ingest(load_corpus()) print('Ready. Ask it anything about the corpus.')

Ask by meaning, not keywords¶

Notice the answers match meaning — no shared-word requirement (the whole point of the course).

In [ ]:

for q in ['why is grass green?',
          'a large ocean animal',
          'strong pain in the side of the head']:
    top = stack.search(q, k=1)[0]
    print(f'Q: {q}')
    print(f'   -> {top["title"]}\n')

for q in ['why is grass green?', 'a large ocean animal', 'strong pain in the side of the head']: top = stack.search(q, k=1)[0] print(f'Q: {q}') print(f' -> {top["title"]}\n')

What you just used¶

• Chunking + ingestion → Chapter 11
• Keyword (BM25) + dense retrieval, fused → Chapters 2, 5, 8
• Embeddings that capture meaning → Chapters 3, 4
• (Optional) re-ranking and RAG answers → Chapters 7, 10

Now start at Chapter 1 and build each piece yourself. Welcome aboard!