oracle-devrel/oracle-ai-developer-hub

Technical resources for AI developers to build applications, agents, and systems using Oracle AI Database and OCI services

What it solves

This repository provides a comprehensive set of technical resources, reference implementations, and tutorials to help developers build production-grade AI applications and agents using the Oracle AI Database and OCI services. It specifically addresses the complexity of managing AI agent memory and data integration by demonstrating how a converged database can replace the need to stitch together separate vector, key-value, and relational stores.

How it works

The hub is organized into several functional areas:

  • Apps: Complete, working reference implementations (e.g., agentic RAG, financial AI agents, and supply chain assistants) that showcase end-to-end architectural patterns.
  • Notebooks: Interactive Jupyter tutorials covering model experimentation, RAG evaluation, and cognitive architectures like Chain of Thought (CoT) and ReAct.
  • Agent Memory: Specialized resources for the oracleagentmemory package, demonstrating how to use Oracle AI Database as a unified memory core for conversation history and durable facts.
  • Guides: Technical documentation and presentations on memory engineering and enterprise agent architecture.
  • Workshops: Guided, hands-on learning paths that progress from basic RAG to advanced memory-augmented agents.

Who it’s for

AI developers and engineers who are building AI agents or RAG systems and want to leverage Oracle's AI infrastructure, specifically those looking for a unified approach to agent memory and data storage.

Highlights

  • Unified Memory Core: Demonstrates using a single database for vector, graph, spatial, and relational queries to power agent memory.
  • Diverse Agent Implementations: Includes examples of multi-agent systems with LangGraph supervisors and specialized assistants.
  • Comprehensive Tooling: Provides integration examples with popular frameworks like LangChain, OpenAI Agents SDK, and Claude Agent SDK.
  • Reference Architectures: Offers detailed guides on "Memory Engineering" and the "2026 agent stack" for enterprise AI.