What this sample helps you learn
This module demonstrates how Oracle AI Database can serve as both **durable agent memory** and an **append-only conversation transcript store** for a LangChain4j assistant. The sample combines:
What this sample demonstrates
Vector Search
Store embeddings and search records by semantic similarity.
Use when users search by meaning or AI answers need grounded records.
JSON
Document-shaped data and SQL/JSON querying inside Oracle AI Database.
Use when records need flexible structure without leaving SQL, indexes, constraints, and transactions.
AI Agents
Runnable AI Agents behavior on Oracle AI Database.
Use when AI Agents needs to be tested against real database behavior.
Highlights
- LangChain4j tools and chat orchestration with gpt-5-nano
- text-embedding-3-small embeddings stored in an Oracle AI Database VECTOR column
- Oracle Text over a native JSON memory document with json_textcontains
- hybrid memory retrieval that fuses semantic similarity with exact text relevance