Hibernate ORM + AI (pgvector) — Spring Boot 3 / Java 21

Minimal, production‑ready starter that combines Hibernate ORM with AI embeddings for semantic search using PostgreSQL + pgvector.

Quick start

1. Start Postgres with pgvector:

   docker compose up -d

2. Run Ollama locally and pull the embedding model:

   ollama pull nomic-embed-text

3. Build & run:

   mvn -q -DskipTests spring-boot:run

API

• POST /api/products — persist + auto-embed
• GET /api/products/search?q=...&k=10 — semantic search
• GET /api/products/search/hybrid?q=...&k=10&wf=0.6&wv=0.4 — hybrid FTS + vector

OpenAI toggle

Set provider to OpenAI in application.yml and provide your API key:

hibai:
  embeddings:
    provider: openai
    model: text-embedding-3-small
    openai:
      base-url: https://api.openai.com/v1
      api-key: ${OPENAI_API_KEY}

Export the key and run:

export OPENAI_API_KEY=sk-...

Flyway migrations

SQL lives under src/main/resources/db/migration and runs automatically on startup.

Testcontainers

Run integration tests against a real Postgres+pgvector:

mvn -q -DskipTests=false test

Demo script

With the app running (and DB up), execute:

scripts/demo.sh

Requires jq for pretty-printing.