I'm a Senior AI and Full-Stack Engineer with 6+ years delivering enterprise-grade systems across AI, cloud, and product development. I specialize in Generative AI and RAG architecture: building document ingestion pipelines, AI chatbots, and intelligent search systems with Azure OpenAI, Azure AI Search, and Semantic Kernel.
At T-Mobile (via IBM) I lead an 18 person cross-platform engineering team building Arrow, T-Mobile's enterprise design system across web, iOS, and Android. Before that I led T-Mobile's enterprise GitHub Copilot rollout, growing adoption to roughly 6,000 developers, building automated access management, and running value stream mapping and training programs across the organization.
Outside of enterprise work, I built the complete React frontend for Gexbot, a real-time options-analytics platform with live WebSocket data streaming that serves a 10,000+ trading community.
I'm a hands-on lead, comfortable going deep on architecture, shipping code, and running teams. I'm currently looking to bring this to a product-focused company where I can build something that compounds over time.
- Description: Built the complete React frontend for gexbot.com, a real-time options-analytics platform for active stock and options traders. Implemented custom, data-dense charts and live gamma-exposure data streaming over WebSockets, and led the state-management refactor from Redux to Zustand for better performance.
- Ideal For: Anyone interested in real-time financial data visualization, WebSocket-driven UIs, or React state management at scale.
- Live Site: gexbot.com
- Description: A keyless, scale-to-zero RAG chat app on Azure: streaming chat with clickable citations and image understanding over your own documents, deployable in one command. Azure Functions (.NET 10) + React, with hybrid (keyword + vector) retrieval and Managed Identity auth (no keys anywhere).
- Ideal For: Developers building context-aware conversational AI, hybrid search, and keyless Azure architectures.
- GitHub Repository: ai-rag-chat-bot
- Description: A production-grade reference Model Context Protocol (MCP) server in TypeScript: one transport-agnostic core serving tools, resources, and prompts over two transports (stdio and Streamable HTTP), backed by a validated data layer, a real test suite, Docker, and CI.
- Ideal For: Developers building or templating their own MCP server with clean, testable architecture.
- GitHub Repository: derekhuynen-mcp-server
- Description: An autonomous Azure pipeline that writes, illustrates, and ships LinkedIn posts on a schedule, with a Next.js dashboard to browse what it has shipped and watch it generate a sample post live.
- Ideal For: Professionals looking to maintain an active digital presence while saving time.
- GitHub Repository: linkedin-ai-auto-poster
- Microsoft Azure AI Engineer Associate (AI-102)
- Microsoft Azure AI Fundamentals (AI-900)
- Microsoft Azure Fundamentals (AZ-900)
- Microsoft Azure Data Fundamentals (DP-900)
- Power Platform Fundamentals (PL-900)
- Email: derek.huynen@gmail.com
- LinkedIn: linkedin.com/in/derekhuynen
- GitHub: github.com/derekhuynen
- Website: derekhuynen.com





