Projects

A journey through the products, platforms, and intelligent systems I've designed, engineered, and brought to production.

AI SaaS · WeCommit

Fundscout

Role: Full-Stack AI Engineer
Timeline: April 2026 – Present
Fundscout Preview

Production-grade AI platform (WeCommit SBIR) that automates federal SBIR/STTR grant discovery and proposal generation for small businesses. Features persistent deep agent conversation memory, real-time multi-user document editing via Yjs CRDTs + Hocuspocus WebSockets, and custom Gemini model routing.

Key Contributions & Technical Impact

  • Architected a 5-layer middleware deep agent stack (Todo, Filesystem, SubAgent, Skills, Memory) using LangGraph and PostgreSQL checkpointing.
  • Built custom GeminiRoutingModel supporting runtime Gemini 2.5 Pro vs Flash model selection and dynamic thinking budgets.
  • Engineered custom streaming protocols mapping LangGraph event streams to Vercel AI SDK DataStreamContext (<3s for 50KB proposals).
  • Integrated real-time collaborative document editing utilizing Yjs CRDT + Hocuspocus WebSockets (<100ms sync latency).
  • Developed custom Tiptap 3 rich-text editor extensions for inline artifact embeds, code rendering, and real-time AI content injection.
  • Resolved async PostgreSQL pooling bottlenecks by implementing automated retry states with exponential backoff, raising uptime to 99.2%.
React 19TypeScriptTiptap 3FastAPILangGraphGemini AIPostgreSQLYjsHocuspocusAlembicPlaywright
2026 – PresentAI SaaS
2026 – PresentAI SaaS
AI SaaS · WeCommit

StageX

Role: Technical Lead & Core Architect
Timeline: March 2026 – Present
StageX Preview

AI-powered startup execution SaaS platform serving 200+ founders during an 8-week accelerator program. Integrates dynamic AI planning with LangGraph agent co-founders, a visual React Flow canvas, real-time streaming, and stateless mentor coordination loops. Deployed live on https://stagex.kr.

Key Contributions & Technical Impact

  • Engineered dynamic, multi-agent SaaS orchestration pipelines using LangGraph and the Vercel AI SDK.
  • Designed a 4-node deep research agent (SearchQuery → WebSearch → Synthesis → Report) with a pre-deduction credit ledger system.
  • Implemented real-time roadmap generation with progressive streaming using Vercel AI SDK (streamObject), rendering large plans in <2s.
  • Developed an interactive tree layout engine using React Flow for visual drag-drop roadmap synchronization with Prisma DB.
  • Engineered token-based mentor invitation/auth system (httpOnly cookies) with dynamic cohort filtering and dashboard feedback UI.
  • Integrated Paddle subscription checking and transactional checkouts with idempotent event webhook handlers.
Next.js 14React 18TypeScriptLangGraphReact FlowPrismaPostgreSQLPaddleNextAuthVercel
AI SaaS · WeCommit

GenDoc (젠닥)

Role: Lead Full-Stack UI & Integrations Engineer
Timeline: March 2025 – January 2026
GenDoc (젠닥) Preview

AI-powered government grant automation and enterprise document collaboration platform designed for startups and SMBs. Streamlines grant discovery, automates proposal drafting matching complex rubrics, and hosts a secure, collaborative rich-text office editor with full revision rollbacks.

Key Contributions & Technical Impact

  • Engineered a premium, responsive React workspace using Radix UI accessible primitives and Tailwind CSS v4, with elegant Framer Motion transitions.
  • Architected and implemented a fully compliant Express.js WOPI Host integrated with AWS S3, streaming active office documents securely into Collabora iframe editors.
  • Developed concurrency collision locking using custom timestamp headers to coordinate multi-user saves and auto-merge actions.
  • Constructed an automatic S3 snapshot timeline component, allowing founders to review previous draft histories and restore versions with zero data loss.
  • Managed security integrations with Better Auth, including Naver/Kakao OAuth systems and Drizzle DDL database schemas on PostgreSQL.
  • Orchestrated multi-stage Docker configurations and automated deployments to Fly.io Tokyo region for optimal latency (<50ms in Korea) and Vercel staging previews.
ReactExpress.jsTailwind CSS v4Better AuthAWS S3Drizzle ORMPostgreSQLDockerFly.ioVercel
2025 – 2026AI SaaS
2024AI
AI · Image Generation & Community Platform

CROSS-E

Role: Full Stack Engineer
Timeline: August 2024 – September 2024
CROSS-E Preview

A creative showcase utilizing advanced generative AI models to convert text prompts into high-fidelity custom images. Integrates a vibrant community feed, real-time social sharing, search index filters, and instant media downloads.

Key Contributions & Technical Impact

  • Integrated OpenAI's DALL-E image generation API into an asynchronous task queue background system.
  • Architected media storage and high-speed distribution networks using AWS S3 and AWS CloudFront CDN cache.
  • Developed a lightweight, high-performance community feed with MongoDB and Express.js to scale concurrent requests.
React.jsNode.jsExpress.jsMongoDBDALL-EAWS S3AWS CloudFront
Hardware · Cryptography & Networks

Raspberry Pi Cloud Storage

Role: Lead Security Engineer
Timeline: January 2023 – April 2023
Raspberry Pi Cloud Storage Preview

A self-hosted personal storage cloud solution constructed on Raspberry Pi hardware to serve as a private, secure, and cost-effective alternative to public commercial cloud services.

Key Contributions & Technical Impact

  • Configured robust multi-layer cryptographic partitions utilizing VeraCrypt, OpenSSL, and custom Unix scripts.
  • Optimized light local network topologies with dynamic DNS and secure Apache servers behind a self-administered MySQL DB.
  • Developed custom PHP services to handle automated backup synchronizations and user permission levels.
Raspberry PiApachePHPMySQLOpenSSLVeraCrypt
2023Hardware
2022Full Stack
Full Stack · Cloud & DevOps

Inventory Management System

Role: Lead Engineer — DevOps & Deployments
Timeline: August 2022 – December 2022
Inventory Management System Preview

An enterprise-grade inventory planning and logistics platform featuring dual-role interfaces for administrators and suppliers, secure authentication layers, and an automated stock replenishment system.

Key Contributions & Technical Impact

  • Containerized backend modules with Docker and orchestrated microservices across multi-node Kubernetes clusters.
  • Architected secure, granular Role-Based Access Control (RBAC) portals for suppliers and organization managers.
  • Deployed scalable cloud infrastructure on IBM Cloud backed by automated GitHub Actions CI/CD pipelines.
PythonIBM CloudDockerKubernetesJavaScriptHTML/CSS