Skip to content
View dunkinfrunkin's full-sized avatar
:shipit:
Parallelizing…
:shipit:
Parallelizing…

Block or report dunkinfrunkin

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
dunkinfrunkin/README.md

Hi, I'm Frank 👋

📍 Irvine, CA | 🏠 SSWE @ Zillow | 🤖 Agentic Engineer | 🏗️ Building the blueprints for autonomous AI

TypeScript JavaScript React React Native Go Python Node.js Claude CLI Codex macOS

Senior Software Engineer at Zillow by day, agentic engineer by night. Neck-deep in the agent stack cataloging the patterns that make LLM-powered systems actually work in production. Less vibes, more blueprints.


🚀 Current Projects

Project What it does
🌐 openagenthub A pattern catalog for architecting LLM-powered autonomous systems agents that reason, use tools, remember, and self-correct. Framework-agnostic, model-agnostic, production-tested.
🧠 openmem Deterministic memory engine for AI agents. Because agents that forget are just expensive autocomplete.
🌊 drift Fast, zero-dependency database migration tool. Everything Flyway paywalls undo, dry-run, cherry-pick, drift detection, schema diff is free and open source.
🐱 mdcat Terminal pager for Markdown. Full colour, syntax highlighting, incremental search, and mouse support zero config.
🔍 uncanny Offline AI-generated text detector. Sentence-level scoring, multiple detection methods, zero API calls. Privacy-first and CLI-native.

🔭 What I'm Working On

  • Mapping the agent landscape Turning scattered tribal knowledge about AI agents into structured, reusable architecture patterns
  • Memory systems for agents Building the primitives that let AI systems actually learn and retain context
  • Bridging demo → production Most agent demos break in prod. I'm documenting why, and how to fix it.

🧭 Philosophy

"Patterns over frameworks." I believe the best way to build reliable AI agents isn't picking the right library; it's understanding the underlying architecture patterns. Frameworks come and go. Good architecture compounds.


⚡ Random Facts

  • 🍩 The username is exactly what you think it is — but plot twist, I'm a tea guy now
  • 🤖 Runs multiple Claude instances like a one-man agent swarm
  • 📐 Believes every AI system is just a graph of patterns waiting to be named
  • 🎓 CS fundamentals enthusiast — from data structures to networking to AI
  • 🍵 Powered by tea and the conviction that agents will eat software
  • 📱 Loves building mobile apps just as much as shipping backend systems

📊 GitHub Activity

GitHub Activity Graph


🤝 Connect

GitHub X (Twitter) LinkedIn openagenthub


"The best agent architectures aren't invented — they're discovered, documented, and shared."

Popular repositories Loading

  1. OpenMem OpenMem Public

    Python 2

  2. drift drift Public

    Go 1

  3. mdcat mdcat Public

    JavaScript 1

  4. cs111-IntroToCS cs111-IntroToCS Public

    Computer Science 111 - Introduction to Computer Science Labs, Exercises, and Final

    Java

  5. cs381-DataCommunicationsNetwork cs381-DataCommunicationsNetwork Public

    Computer Science 111 - Data Communications and Network Labs and Exercises

    Python

  6. cpsc481AI cpsc481AI Public

    Computer Science 481 - Artificial Intellegience

    CMake