Welcome to DevRel Labs, a collection of AI solutions and tools built with Oracle technologies. This repository contains practical implementations, demos, and guides for various AI use cases.
apps/: Deployable applications, automation scripts, and infrastructure-as-code projects.workshops/: Step-by-step guides, LiveLabs content, and educational materials.notebooks/: Reserved for notebook-driven explorations and tutorials (currently minimal).
Building AI Agents with Model Context Protocol (MCP) and Oracle Database 26ai
- MCP Integration: Secure tool calling between LLMs and Oracle Database
- Real-time Data Access: Live business data integration without ETL
- Vector RAG: Native vector search capabilities
- Langflow Integration: Visual workflow builder for AI agents
- Use Cases: Financial advisor, customer service, business intelligence
Talk with Industry-Specific Domain Data and Gain Insights Using Select AI
- Natural Language Queries: Ask questions in plain English
- Domain-Specific Knowledge: Industry-tailored insights and analysis
- No SQL Required: Democratized data access for business users
- Multi-Industry Support: Finance, healthcare, retail, manufacturing
- Use Cases: Business intelligence, compliance, operational analytics
- Agentic RAG - AI RAG in a BOX Demo using Oracle AI Database 26ai and Local LLMs
- PlaneLLM - PlaneLLM integration with Oracle Database
- Neural Networks Hero - Neural network implementations and tutorials
- OCI Vision AI - Computer vision solutions using Oracle Cloud Infrastructure
- Mask Detection - Training and labeling for mask detection models
- OCI Language Translation - Multi-language translation services
- OCI Subtitle Translation - Automated subtitle translation
- OCI CSV-JSON Translation - Data format conversion tools
- NVIDIA NIM on OKE - NVIDIA Inference Microservices on Oracle Container Engine
- Holoscan - Holoscan applications and deployments
- Kubeflow on OKE - Machine learning workflows on Kubernetes
- Data in AI Revolution - Data-driven AI solutions and insights
- RAG in a Box - Retrieval-Augmented Generation solutions
- agentic_rag - Multi-agent RAG system with Chain-of-Thought reasoning using Oracle Database 26ai and local LLMs
- holoscan - Terraform stack for deploying NVIDIA Holoscan on Oracle Linux A10 GPU instances
- langflow-agentic-ai-oracle-mcp-vector-nl2sql - Agentic AI workflows with Langflow, Oracle Database MCP, Vector RAG, and NL2SQL
- langgraph_agent_with_genai - File indexing and conversational search using LangGraph and Oracle Generative AI
- mongo-migration - Migration tools and solutions for moving MongoDB applications to Oracle Database
- nvidia-nim-oke - Deploying NVIDIA Inference Microservices on Oracle Kubernetes Engine with GPU support
- oci-csv-json-translation - Translate specific columns in CSV files or keys in JSON documents using OCI Language
- oci-language-multiple-translation - Bulk translation of multiple documents from OCI Object Storage buckets
- oci-language-translation - Multi-language translation services for documents and text using OCI Language
- oci-subtitle-translation - Transcribe audio files and translate subtitles using OCI Speech and Language services
- OJET (VDOM) - OCI Vision - Oracle JET Virtual DOM application with OCI Vision integration for image analysis
- oracle-mcp-ai-agents - Building AI agents with Model Context Protocol (MCP) and Oracle Database 26ai
- oracle-rag-applications - Building RAG applications with Oracle Database 26ai for document processing and semantic search
- oracle-select-ai-insights - Natural language queries for industry-specific domain data using Oracle Select AI
- planeLLM - Generate bite-sized educational podcasts on any topic using OCI GenAI service
- rag_in_a_box - Containerized RAG system with Oracle Database 23ai and local LLMs deployable via Podman
- ai-meetings - Build web applications using Visual Builder Cloud Service with AI-powered meeting transcription, summarization, and sentiment analysis
- data-in-ai-revolution - Educational workshop on data types, structures, and processing techniques in AI and machine learning
- mask_detection_labeling - Step-by-step guide for creating and labeling a computer vision model to detect mask-wearing states using RoboFlow
- mask_detection_training - Training and improving YOLO-based object detection models for mask detection on Oracle Cloud Infrastructure
- neural_networks_hero - Comprehensive workshop on neural network implementations, training, and inference for computer vision tasks
- Currently empty - Reserved for notebook-driven explorations and tutorials
Each solution includes:
- Comprehensive README with architecture overview
- Quick start guides and prerequisites
- Configuration examples and deployment options
- Use cases and best practices
- Troubleshooting guides
We welcome contributions! Please see individual solution directories for contribution guidelines.
Copyright (c) 2024 Oracle and/or its affiliates.
Licensed under the Universal Permissive License (UPL), Version 1.0.
See LICENSE for more details.
Built with ❤️ by the Oracle DevRel team