Juyoung Kim

Software / Product Engineer · San Francisco, CA · U.S. Citizen

juyoungk23@gmail.com · GitHub · LinkedIn

Download PDF

Skills

Swift/SwiftUI, Swift 6 concurrency, Python, TypeScript, SQL · React, Next.js, Supabase (Postgres), GCP · computer vision, LLM integration (Claude, Gemini) · RealityKit, AR/VR, 3D Gaussian Splatting

Work Experience

San Francisco, CA
  • Founding engineer on the three-person team behind Poppy, a proactive AI assistant (iOS, macOS, watchOS) that connects calendar, email, messages, and health data and surfaces what matters before users ask. Joined pre-launch; shipped the May 2026 public debut covered by TechCrunch and 9to5Mac. Backed by Kindred Ventures.
  • Co-built Pods, Poppy's long-running agentic tasks; own the pods backend day to day and the client that renders agent output (charts, chips, deep links) as native SwiftUI.
  • Extended Poppy across iOS, watchOS, and macOS with cross-device handoff: a suggestion on iPhone or Watch opens the meeting link on the Mac (WatchConnectivity relay into MultipeerConnectivity).
  • Built two-way sync between Poppy's reminder system and Apple Reminders via EventKit, with Poppy's backend as the source of truth: reminders created in either app (or by the assistant) appear in both, offline edits and deletions reconcile on re-sync, and ownership logic ensures only one alarm ever fires.
  • Built the health layer integrating HealthKit and Oura Ring data, deduplicating overlapping Apple Watch and ring signals, and shipped sleep briefings that explain scores with the life events behind them.
  • Set up product analytics from zero (Mixpanel, in-depth dashboards, weekly reports) and shipped a new onboarding animation sequence that analytics showed had ~60% end-to-end completion through OTP verification and a paywall; introduced the codebase's first test target (Swift Testing package, ~30 tests, pre-commit hooks).
San Francisco, CA
  • Designed and ran a high-throughput migration pipeline (NoSQL to SQL) from critical legacy systems to Google Cloud with zero data loss, and refactored Python ETL scripts to cut processing latency by 60%.
  • Led monthly production deployments, coordinating code integration across multiple teams and enforcing SDLC standards.
  • Previously interned (Summer 2022): won 1st place in the company AI Hackathon with a voice-impersonation detector built on Google Vertex AI and audio spectrogram classification.

Technical Projects

Fight analytics from broadcast video (Python, PyTorch, Next.js)
  • Built an end-to-end computer-vision system that turns single-camera MMA broadcasts into verified fight analytics: high-recall strike detection with a human-in-the-loop verification workflow, and a live site with per-fighter reports across 22 fighters.
iOS, Web, visionOS
  • Built a cross-platform app that converts a single 2D photo into a 3D Gaussian Splat via an event-driven serverless GPU pipeline; an iOS App Clip (< 15MB) lets recipients view shared memories from an iMessage link with no install. Live on the web and the App Store.
SwiftUI, iOS, visionOS
  • Launched a subscription immersive streaming app on the iOS App Store and Apple Vision Pro with an engaged user base and broadcast partnerships with Mixed Martial Arts promotions across the United States; extracted the SharePlay synchronization engine into the open-source ImmersiveWatchParty SDK.

Education & Certificates

  • B.S. in Computer Science, UNC Chapel Hill (2023)
  • Google Cloud Professional Cloud Architect | Google Cloud Associate Cloud Engineer