Software Engineer | C++/Linux & Embedded Systems | AI Infrastructure

I'm a software engineer with 6+ years of experience building production C++/Linux software at Hanwha Vision, spanning multi-server video systems, embedded devices, and cross-platform applications. I focus on solving performance and reliability problems with measurable, tested engineering changes.

I'm now pursuing an M.S. in Artificial Intelligence at Santa Clara University, extending that systems background through merged AI inference open-source contributions and hands-on AI systems projects.

🎯 Seeking Summer 2027 internships: Software Engineering · C++/Embedded Systems · AI Infrastructure / ML Systems
📍 San Jose, California · San Francisco Bay Area
🔗 Portfolio · LinkedIn · Email

• Production systems: Developed modular C++/Qt 6/QML components for a multi-server video management platform supporting thousands of camera and server connections, including REST-backed device management.
• Performance optimization: Reduced firmware-information load/reset latency 67% (1,174 ms → 386 ms) using parallel execution threads; cut C++ build time 13.9% with precompiled headers, validated across five benchmark runs per configuration.
• Engineering at team scale: Managed Git branching, reviews, and release workflows for a 50-engineer team, supporting biweekly releases and urgent hotfixes.
• Embedded and cross-platform development: Integrated and hardware-tested NVR LTE modem functionality in approximately three weeks; built software for Windows, macOS, and Ubuntu and automated embedded SoC toolchain setup with Docker.

Qualcomm GenieX — AI inference runtime

• PR #1564 · C++ / VLM memory allocation: Changed logits allocation from per-token rows to one row per sequence, avoiding unused memory reservations.
• PR #1577 · Python / streaming reliability: Fixed cancellation behavior that could emit SDK-rejected tokens and leave streamed output inconsistent with the final result; addressed prefix-reuse issues.

AWS SageMaker HyperPod CLI — developer tooling

• PR #471 · CLI correctness and regression tests: Aligned CloudFormation stack output, CLI help, and SDK documentation, with regression tests.

• CardGuard · Collaborative AI Security: Worked in a five-person hackathon team on prompt-injection defenses, rogue-node safeguards, and agent coordination for privacy-preserving payment-risk analysis.
• ONVIF Professional Plugin · Developer Tooling: Built a Claude Code plugin for network-camera protocol diagnostics, device-profile analysis, and conformance workflows.
• AI Agent Security Labs · Security Workbench: Implemented tool-call policies and evidence auditing; 234 backend tests passed using synthetic data.
• RippleBrain · Enterprise Change-Impact Agent: Built a FastAPI/RAG application that connects retrieved evidence to change-impact analysis and human review and approval.

• Languages: C++, C, Python, SQL, JavaScript, TypeScript
• Systems & engineering: Linux, embedded systems, Qt 6/QML, multithreading, Docker, Git/GitHub Actions, CI/CD, REST APIs, Windows, macOS
• AI & backend: Machine Learning, LLM inference, RAG, AI agents, FastAPI, Next.js

• Santa Clara University — M.S. in Artificial Intelligence (expected June 2028)
• Hongik University — B.S. in Computer Engineering (2020)

More about my engineering work: Interactive portfolio