John Yoon

AI/ML engineering leader building reliable GenAI systems, research workflows, and mission-oriented software.

Interested in AI systems that improve learning, decision support, ministry operations, and human formation.

John Yoon

What I do

Reliable AI systems for work that needs judgment.

Featured Case Studies

Production systems, confidential constraints, public signal.

Proprietary work cannot show code, so these summaries focus on the problem, architecture, impact, and operating constraints.

Support automation and expert workflows at production scale.

Intuit GenAI support orchestration

Problem
Route complex customer-support work through GenAI systems without losing reliability, escalation control, or human trust.
Architecture
Orchestration layers around LLM calls, retrieval, guardrails, evaluation loops, expert matching, and production ML services.
Impact
Improved the path from customer question to useful answer while keeping quality, safety, and operational constraints visible.
Constraints
Proprietary systems; public summary focuses on problem framing, architecture patterns, and reliability tradeoffs.
Fraud, risk, and production data systems.

Capital One fraud/ML systems

Problem
Detect and mitigate risky activity in high-volume financial workflows where false positives, latency, and auditability matter.
Architecture
Model development, feature/data pipelines, experimentation, monitoring, and production decision-support systems.
Impact
Built ML capabilities for risk operations where model behavior had to be measurable, explainable, and production-ready.
Constraints
Financial-services work is confidential; public summary keeps customer, model, and platform details abstract.
Open-source tools for research and local-first developer workflows.

Public tools: research-workbench, activitywatch-calendar-export, tmx, symlink-logger

Problem
Make everyday research, time analysis, terminal sessions, and filesystem migration safer to inspect and automate.
Architecture
Small CLIs and local automation around structured project state, calendar exports, tmux/session workflows, and FUSE-backed access logging.
Impact
Public code demonstrates practical systems taste: narrow interfaces, repeatable workflows, and tools built for real personal use.
Constraints
Designed for local workflows first, with privacy-preserving examples and public-safe documentation.

Substack & GitHub

Latest visible output

Built daily from public GitHub metadata and Substack RSS.

Substack Writing

Public essays

Longer-form public writing will appear here when Substack RSS returns published posts.
GitHub Swift

endel-focus-menubar

macOS menu bar timer for Endel focus sessions with TaskForge-compatible logging.
GitHub Python

idle-maintenance

A macOS maintenance suggestion system that prompts users with actionable tasks during idle time
GitHub Shell

tmx

T-max: a Git-inspired tmux workflow CLI
GitHub Python

iterm-session-restore

Restore recently closed iTerm2 tabs and panes, with optional agent session resume hooks.

Activity

Public conversations

Official platform embeds, with no private API tokens.

Selected Work

Tools worth inspecting

Go / systems

symlink-logger

FUSE-backed CLI for safely retiring directory symlinks by logging compatibility-path access.

 Developer tools

aicommit-split

Deterministically split working-tree changes into semantic commit groups before running aicommit.

 Research workflows

research-workbench

CLI tools for organizing research projects, sources, and Readwise workflows.

 Local-first automation

activitywatch-calendar-export

Export ActivityWatch timelines to Obsidian summaries and calendar feeds.

CV

AI engineering, research workflows, and mission-context product building

Senior AI/ML engineering GenAI orchestration M.Div. candidate 3 U.S. patents Founder

Engineering

  • Senior Machine Learning Engineer, Intuit — production GenAI orchestration, AI guardrails, support automation, ML systems.
  • Machine Learning Engineer / Data Scientist, Capital One — fraud/risk ML, model development, production data systems.
  • Member of Technical Staff at Nutanix; automation and release infrastructure.

Mission & Research

  • M.Div. candidate at Southwestern Baptist Theological Seminary, expected 2026.
  • Interested in AI systems that improve learning, decision support, ministry operations, and human formation.
  • Founder of Wikicourse, a collaborative lecture and learning platform.