Nguyen Ngoc Quy

Software Engineer | C++ Systems, AI Infrastructure & Backend Tooling

[email protected] +84 918 211 296 Hanoi, Vietnam quynn.dev

English · Tiếng Việt

Summary

Software Engineer with 5+ years of experience in high-performance C++ systems, backend tooling, and local AI infrastructure. Specialized in containerized vLLM serving, agentic routing gateways, tool-calling workflows, and large-scale server stress testing. Proven track record of optimizing latency, concurrency, and developer productivity across distributed teams.

Work Experience

Game Programmer (C++ Systems & AI Infrastructure) 2021-03-01 — Present
Gameloft
  • Local LLM Infrastructure & Serving (Jun 2026–Present): Deployed containerized LLM inference engines on multi-GPU nodes; optimized vLLM memory allocation, KV-cache, context length, concurrency, and speculative decoding via reproducible stress and tool-use benchmarks.
  • LLM Gateway & Agentic Routing: Engineered an Anthropic/OpenAI-compatible AI Gateway in TypeScript to route agentic developer tools (Claude Code, OpenCode) across local vLLM instances and cloud models with streaming protocol translation, authentication, and PostgreSQL usage metering.
  • Automated Crash Analysis Pipeline: Architected an AI-assisted Linux core-dump diagnostic platform (S3 → WSL/GDB with archived symbols) that retrieves contextual stack traces, groups related crashes, and leverages LLMs for automated root-cause analysis and patch proposals via a PostgreSQL dashboard.
  • Automated SVN Code Review System: Developed a model-agnostic code review platform for SVN repositories featuring scheduled LLM evaluations, finding classification, verdict tracking, and notification workflows backed by PostgreSQL.
  • Backend Load & Stress Testing: Built an automated testing platform simulating thousands of concurrent connection players to stress-test game backend servers and event APIs prior to production deployment.
  • Core C++ & Client-Server Gameplay: Developed high-performance C++ gameplay systems and features across both client and server, contributing to Android platform porting and optimization.
  • Cross-Stack & SDK Delivery: Delivered core features across ActionScript legacy titles and Unreal Engine 5 migration projects, integrating third-party backend SDKs including Netflix and regional distribution platforms.
  • CI/CD Infrastructure: Designed and maintained Jenkins build and deployment pipelines for Sniper Fury, improving release stability and operational efficiency.
  • Technical Leadership: Led technical workshops and programming contests while mentoring junior engineers and standardizing technical documentation in fluent business English.

Skills

Core Systems & Backend Engineering: C++, Python, Node.js, REST APIs, PostgreSQL, Android Platform Development, Linux, GDB, Docker, Podman
AI Engineering & LLM Infrastructure: Local LLM Inference, vLLM, Model Serving Optimization, LLM Gateways, Tool Calling & Function Calling, Agentic Workflows, Claude Code, OpenCode, Claude Agent SDK
DevOps & Professional Tools: Java, Unreal Engine 5, Jenkins CI/CD, Git / GitFlow, SVN, Jira

Education

Hanoi University of Science and Technology 2014-09-01 — 2021-10-01
BS in Global ICT (Global Engineer Program in Information and Communication Technology)

Languages

English: Full Professional Proficiency
Vietnamese: Native

Download as PDF