feat: modernize ner-powered-search notebook to Pinecone SDK v8

[jhamon]

Summary

This PR modernizes the ner-powered-search.ipynb notebook to use Pinecone SDK v8, updating it from the legacy pinecone-client package to the modern pinecone package and aligning it with current SDK patterns and repository standards.

Problem

The notebook was using the deprecated pinecone-client==3.1.0 package, which is no longer maintained. This creates confusion for users and doesn't reflect current best practices for working with Pinecone.

Solution

Updated the notebook to use Pinecone SDK v8 with the following changes:

• Updated package installation: Changed from pinecone-client==3.1.0 to pinecone (unpinned to get latest v8)
• Modernized SDK imports: Updated from legacy pinecone module to modern Pinecone client class
• Standard API key handling: Implemented the standard pattern using os.environ.get() with getpass() fallback instead of hardcoded placeholder
• Added Prerequisites section: Included clear listing of required packages and API keys for better user guidance
• Code quality improvements: Applied repository formatting standards (ruff) and stripped cell outputs (nbstripout)

Key Concepts

• ServerlessSpec: The notebook already used ServerlessSpec for index creation, which is the correct v8 pattern
• Pinecone client initialization: Uses modern Pinecone(api_key=api_key) pattern
• API key security: Follows best practice of environment variables with interactive fallback

Breaking Changes

None. The notebook functionality remains the same; only the underlying SDK package and initialization patterns have been updated.

Example Usage

Users can now run the notebook with the latest Pinecone SDK v8:

# Install with modern package name
!pip install sentence_transformers pinecone datasets -qU

# Initialize with modern pattern
from pinecone import Pinecone
api_key = os.environ.get("PINECONE_API_KEY") or getpass("Enter your Pinecone API key: ")
pc = Pinecone(api_key=api_key)

Testing

• ✅ Pre-commit hooks pass (ruff formatting, nbstripout, codespell)
• ✅ Notebook structure validated against NOTEBOOK_REVIEW_TEMPLATE.md
• ✅ All SDK v8 patterns correctly implemented
• ✅ Prerequisites section added
• ✅ API key handling follows standard pattern

---

Closes: SDK-161

🤖 Generated with Claude Code

---

Note

Low Risk
Low risk: documentation/demo notebook changes only; main impact is runtime dependency/API usage updates for Pinecone initialization and index creation.

Overview
Updates ner-powered-search.ipynb to use the modern pinecone package (SDK v8) instead of the legacy pinned pinecone-client dependency, while keeping the NER + vector-search workflow the same.

Adds a Prerequisites section, switches API key handling to PINECONE_API_KEY with an interactive getpass() fallback, and reformats/cleans the notebook by stripping execution outputs and applying small style/formatting tweaks (e.g., compact imports and argument formatting).

^{Written by Cursor Bugbot for commit 71c7d2c. This will update automatically on new commits. Configure here.}

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.}

[cursor]

Query uses positional argument instead of keyword argument

Low Severity

The index.query() call passes the vector xq as a positional argument, but all other notebooks in this repository use vector=xq as a named keyword argument. This violates the review rule: "Prefer to use named keyword arguments over positional arguments." For a PR claiming to modernize to SDK v8 patterns, the query call should be index.query(vector=xq, top_k=10, ...) to match the repository's established pattern and improve code clarity.