Kiyoung So — Full-Stack & AI Engineer

What I do

Centered on AI & agents — standing on deep full-stack, platform, and real-time CV/ML experience.

AI & agent tooling

Drive parallel AI-agent sessions from your phone — a supervisor spawns workers that collaborate and report back. Local-first.

Full-stack product

React · React Native · Next · Node · NestJS · FastAPI · Spring — 0→1 to production.

Platform & infra

AWS, Docker, CI/CD, real-time pipelines, licensing, zero-downtime deploys.

Real-time CV/ML

Webcam gaze tracking in C/WASM (SIMD) — 6-DOF pose, fixation detection. Live below.

Live demo

The deepest layer, proving the rest: real-time CV/ML compiled to WebAssembly, running on your own webcam. Read the passage for 12 seconds and watch your gaze path replay. No install, nothing uploaded.

Most people think they read in smooth lines. In fact your eyes move in quick jumps — saccades — pausing only at brief fixations a few times a second, and you are effectively blind in between.

Webcam eye-tracking, compiled to WebAssembly — runs entirely on your device.

~30 fps · webcam-bound C/WASM · SIMD SharedArrayBuffer zero-copy 6-DOF pose · 1€ filter · fixation (dispersion)

Also running in production for real clients.

Selected work

R_workstation — multi-agent dev platform

AI / agents

solo · daily driver for 3 devs · local-first

From your phone, say what you want → a supervisor wakes parallel (Claude Code) sessions that work together and report back
Hono + socket.io + SQLite backend · React PWA · runs on your own machine or a rented server
TOTP auth · Tailscale Funnel · child-process lifecycle management

Local-first: zero data leaves the machine.

Commercial eye-tracking analytics platform

full-stack + infra

end-to-end owner · led a 4-person team · built the core myself

Hardware→browser bridge: Tobii C++ SDK → Node native addon → Electron → WebSocket → React
HTML5 Canvas gaze analytics (replay, heatmaps, fixations, scanpath, AOI) + behavioral-scoring ML
Full AWS infra: CI/CD, Redis, OAuth, monitoring, zero-downtime deploys

Led the team while architecting and building the core end-to-end.

Browser webcam gaze tracking (CR_wasm)

real-time CV/ML

solo · ~30 fps in-browser · live in the demo above

Real-time CV/ML compiled to C/WASM (Emscripten, SIMD) with a SharedArrayBuffer zero-copy frame path
6-DOF head pose (solvePnP), 1€ filter smoothing, fixation detection (dispersion threshold)
HMAC + ECDSA licensing · S3 / CloudFront distribution

Owned every layer: CV math → WASM build chain → browser runtime → cloud.

Tobii 5L native integration

hardware → cloud

solo · full system boundary

C++ SDK → Node native addon → Electron, including the full build chain
IPC / WebSocket real-time streaming, live visualization, data analysis
Cloud sync + licensing

Designed the entire hardware → cloud boundary alone.

And a decade more — a Unity/Vuforia AR reading app (point a phone at a book → characters appear over the words), AWS Alexa voice skills, React Native / Android apps, and many others over 11 years.

How I work

Remote is my only hard requirement — and I'm fully timezone-flexible: I work your team's hours, nights included, anywhere. Open to Founding / Staff Engineer roles and contract / EOR engagements with teams that move on results.

My edge isn't breadth for its own sake — it's breadth that stands on real depth in every layer. If you're building AI/agent tooling, full-stack products, or real-time/edge systems and want someone who can own the whole thing — let's talk.

ssogi@ssogi.com