Skip to content
View johnson2500's full-sized avatar

Block or report johnson2500

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
johnson2500/README.md

Hi there, I'm Ryan but people call me RJ! πŸ‘‹ πŸ€–

I am a Principal Software Engineer at Red Hat with over a decade of experience building scalable systems, spinning up Kubernetes clusters, and teaching AI agents how to do my job (just kidding... not really tho).

Right now, I spend my days orchestrating the Llama Stack, building agentic quickstarts for OpenShift AI, and convincing autonomous agents to cooperate.


πŸš€ What I'm Up To

  • 🧠 Red Hatting: Building cutting-edge AI frameworks, RAG pipelines on Kubernetes, and agentic workflows.
  • 🦜 Agent Wrangler: Building proof-of-concepts with CrewAI, LangChain, LangGraph, and vLLM.
  • πŸ“¦ Production AI: Bridging the gap between raw LLMs and production-grade ML Ops.

πŸ› οΈ My Tech Stack

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚   AI, LLMs & Agents: Llama Stack | CrewAI | LangChain | LangGraph | Llama Index        β”‚
β”‚                      vLLM | Llama.cpp | Ollama | OpenAI | Claude | OpenClaw            β”‚
β”‚                      Semantic Routing | Agentic Frameworks | RAG | Graph RAG          β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚   Vector & DBs:      PGVector | Pinecone | ChromaDB | Postgres | Mongo | MySQL         β”‚
β”‚                      Snowflake | BigQuery | RDS | Redis Vector Store                   β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚   Cloud, K8s & OS:   Kubernetes | Red Hat OpenShift | AWS | GCP | Firebase             β”‚
β”‚                      Red Hat Linux | Docker                                            β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚   Backend & APIs:    Node.js | NestJS | TypeScript | Python | Fast API | Express       β”‚
β”‚                      HapiJS | Fast MCP | Schema & API Design | API Routing | Caching   β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚   Frontend & UI:     ReactJS | NextJS | HTML5 | CSS3 | jQuery Mobile                   β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
  • Languages: JavaScript, TypeScript, Python, SQL
  • Frameworks & Runtimes: NestJS, Next.js, Express, Fast API, React
  • Infrastructure & Ops: Kubernetes, Red Hat OpenShift, Docker, AWS, GCP, Linux

πŸŽ₯ The Backstory (A Quick TL;DR)

  • Origin Story: It all started with an Electrical Engineering degree at UMass Dartmouth and looking at flashing router lights at State Street back in 2015.
  • Way Back: Built APIs, data pipelines, and microservices for heavy-hitters like 3Play Media, VideoAmp, Disney, and 20th Century Fox.
  • Before: Migrated media infrastructures to AWS and built custom RAG systems for journalists/lawyers at Bongo Media.
  • Now: Teaching Red Hat OpenShift AI how to think via agentic frameworks.

🀝 Connect with Me

const ryan = {
  status: "Deploying AI agents with Openshift AI",
  coffeeConsumption: "High",
  askMeAbout: ["Llama Stack", "Kubernetes", "Why my LangGraph agent is looping"]
};

Pinned Loading

  1. ai-architecture-charts ai-architecture-charts Public

    Forked from rh-ai-quickstart/ai-architecture-charts

    This repository contains all the helm charts to deploy LLM service, Llama Stack server, configuring pipeline server, minio, pgvector

    Python

  2. ai-virtual-agent ai-virtual-agent Public

    Forked from rh-ai-quickstart/ai-virtual-agent

    Deploy AI agents with knowledge bases and tools on OpenShift

    Python

  3. RAG RAG Public

    Forked from rh-ai-quickstart/RAG

    A QuickStart that allows users to experiment with retrieval-augmented generation (RAG) in a streamlined chat environment.

    Python

  4. rh-ai-quickstart/nvidia-blueprint-retail-spending rh-ai-quickstart/nvidia-blueprint-retail-spending Public

    Repository for code related to deploying Nvidia's Retail Spending Blueprint to run in Openshift.

    Jupyter Notebook 1

  5. Billing-extraction-with-GroundX Billing-extraction-with-GroundX Public

    Forked from rh-ai-quickstart/Billing-extraction-with-GroundX

    AI quickstart to leverage GroundX to extract billing information from a billing statement.

    Python

  6. rh-ai-quickstart/Fraud-Detection-data-versioning-with-lakeFS rh-ai-quickstart/Fraud-Detection-data-versioning-with-lakeFS Public

    AI quickstart to show how lakeFS can be used for data versioning in a fraud detection application.

    Makefile 5 2