AI-Powered Document Search with Oracle and ONNX Embeddings for Recruiting

How it works

Create a user for doing Hybrid Search. Clear Existing Data, if present. Add Documents into the table. Create a hybrid index. Run Semantic search on Documents table for "prioritize teamwork and leadership experience". Run Hybrid search for the text inputted in Chat interface.

Setup Steps

Download the ONNX model

all_MiniLM_L12_v2_augmented.zip

Extract the ZIP file on the database server into a directory, for example /opt/oracle/onnx. After extraction, the folder contents should look like:

bash-4.4$ pwd /opt/oracle/onnx bash-4.4$ ls all_MiniLM_L12_v2.onnx

Connect as SYSDBA and create the DBA user

-- Create DBA user CREATE USER app_admin IDENTIFIED BY "StrongPassword123" DEFAULT TABLESPACE users TEMPORARY TABLESPACE temp QUOTA UNLIMITED ON users;

-- Grant privileges GRANT DBA TO app_admin; GRANT CREATE TABLESPACE, ALTER TABLESPACE, DROP TABLESPACE TO app_admin;

Create n8n Oracle DB credentials hybridsearchuser → for hybrid search operations dbadocuser → for DBA setup (user and tablespace creation)

Run the workflow

Click the manual Trigger It displays Pure semantic search results

Enter search text in Chat interface It displays results for vector and keyword search.

Note The workflow currently creates the hybrid search user, docuser with the password visible in plain text inside the n8n Execute SQL node. For better security, consider performing the user creation manually outside n8n. Oracle 23ai or 26ai Database has to be used`

Reference Hybrid Search End-End Example