Oracle open-sourced an AI developer hub that's worth studying seriously.
A technical repository with reference implementations, benchmarked notebooks, and guided workshops, built around one idea: Oracle AI Database as a unified memory core for AI agents.
If you're learning how to build agents beyond the vector DB tutorial, this is the repository to go deep on.
Here's what's inside:
📱 Application reference implementations:
🔹 finance-ai-agent-demo - financial AI agent running vector, graph, spatial, and relational queries on a single Oracle AI Database simultaneously
🔹 agentic_rag - multi-agent RAG with Chain of Thought, processing PDFs, websites, and code repos
🔹 FitTracker - fitness platform built live during a webinar using Oracle 26ai JSON Duality Views FastAPI Redis
🔹 tanstack-shoestore - natural language database queries using Oracle 26ai Select AI
🔹 oci-generative-ai-jet-ui - full-stack with Oracle JET, Kubernetes, and Terraform
📓 Notebooks that go deeper than most tutorials:
🔹 11 cognitive architectures in one interactive demo - CoT, ToT, ReAct, Self-Reflection, Least-to-Most, Decomposed Prompting
🔹 Vector keyword hybrid search in a single SQL query - not three calls to three systems
🔹 6 types of persistent memory implemented and compared
🔹 Filesystem vs database agent memory - compared directly
🔹 F1 Miami 2026 GP strategy using real FastF1 data - SQL, hybrid search, JSON documents, and property graph all in one Oracle 26ai database
🧠 Agent Memory package (OAMP):
🔹 One converged engine replacing vector DB key-value store relational store
🔹 Benchmarked against naive flat-history memory - 80 turns, 3 agent variants, token cost latency quality measured
🔹 Three core primitives: users/agents, memories, threads
🔹 End-to-end examples for OpenAI Agents SDK, Claude Agent SDK, and LangGraph
🔹 Deep Research Agent - genome exploration with Tavily OAMP for durable findings across sessions
🔹 Mortgage Approval in LangGraph - OAMP persists applicant data so failed runs resume where they stopped
🏫 3 workshops, each self-contained with Codespaces environment and Oracle AI Database pre-configured:
🔹 Information Retrieval to RAG - 5 retrieval strategies over 200 ArXiv papers
🔹 From RAG to Agents - retrieval as tools, persistent session memory
🔹 Agent Memory - 6 memory types, context engineering, with vs without memory compared
Multicloud samples: AWS, Azure, Google Cloud, MongoDB API.
Presented at DevWeek SF 2026 and the DeepLearning[.]AI AI Developer Conference, April 2026.
Open source.