Uploaded June 2026 | Updated September 2026, 3 weeks ago

---

Learn how to build semantic search over real GitHub issues using Oracle AI Database 26ai, Free SQL, and LangChain Oracle DB.
In this hands-on tutorial, Kesha Williams shows how to move beyond keyword search by storing GitHub issues as vectors in Oracle AI Database, running similarity search, filtering results with metadata, and dropping down into native vector SQL when you need more control.

Repo link: github.com/oracle-devrel/oracle-ai-developer-hub/tree/main/workshops/semantic-search-github-issues

Chapters:
00:00 Why keyword search misses duplicate issues
00:17 What you’ll build
00:37 Oracle AI Database 26ai, Free SQL, and Codespaces
01:17 Set up Free SQL credentials
01:39 Pull real GitHub issue data
02:18 Connect from the notebook
03:29 Create LangChain documents
03:52 Load the embedding model
04:28 Store GitHub issues as vectors
04:51 Run semantic similarity search
06:26 Compare with keyword search
07:28 Add metadata filtering
08:33 Use native Oracle vector SQL
10:01 Run it on your own repository

#OracleAIDatabase #VectorSearch #SemanticSearch #LangChain #Python #GitHub #AIDevelopment