I design Agentic AI workflows, RAG pipelines, and low-latency microservices that bridge advanced ML with measurable enterprise ROI. For a decade I've been the person teams call when a system has to stay up, ship fast, and still pay for itself.
- 🧠 Now — Consulting AI Architect · Agentic AI + RAG blueprints across GCP & Azure
- 🛠️ Recently — Led AI inference optimization at CADDi: cut latency 60%, saved ~$180K/year
- 🎯 Open to — Senior / Principal / Staff AI Architect roles · Education & Healthcare collabs
- 🎓 Carnegie Mellon × Van Lang · BEng, Computer Software Engineering (2010–2016)
- 🎤 Off keyboard — I sing. Ask me about the setlist.
|
manufacturing drawings indexed & queryable |
AI inference latency reduction |
annual cloud infrastructure saved |
production uptime SLA across 12 services |
|
RAG document chunks in vector stores |
concurrent enterprise learners served |
enterprise cloud spend via FinOps governance |
faster candidate-to-role matching (14d → 3d) |
2026 ▸ Independent Consulting · Consulting AI Architect · Agentic AI + RAG blueprints (GCP, Azure)
2023 ▸ CADDi · Senior Full Stack Engineer · 3M+ drawings · 60%↓ latency · $180K/yr saved
2020 ▸ Kydon Group 🇸🇬 · Front End Lead · 10K+ learners · sub-200ms · MCP-style APIs
2019 ▸ Vincere.io · Senior Web Developer · 2K+ firms · 5× faster placement · 99.9% SLA
2018 ▸ Spirit Labs · Full Stack / Distributed · Formal verification · 95%+ TDD coverage
2017 ▸ ekino Vietnam · Frontend Developer · SPAs for 500K+ MAU · 45%↓ page latency
2016 ▸ Ministry of Natural Resources · Jr Web Dev / Data Systems · 50K+ datasets · 12 provincial nodes
🛡️ dom-defenderStack: React · TypeScript · Vite · Tailwind |
Cross-platform Tauri desktop app + CLI for managing git aliases. 270+ curated aliases, 229 tests, Rust core with a React/TS shell. The tool I wish existed when I was onboarding engineers across 6 repos.
Stack: Rust · Tauri · React · TypeScript |
💌 wife-cvStack: TypeScript · Vite · Node.js |
Long-form notes on agentic AI, RAG architecture, inference economics, and the occasional rant about micro-frontends — on Medium.
- [Prompts are code. Mine has a regression harness.](https://medium.com/@zintaen/prompts-are-code-mine-has-a-regression-harness-b7e6c78310f2?source=rss-a77250ac42f4------2) Apr 22, 2026- [I Built a Desktop App to Fix the Messiest Part of My Git Workflow](https://medium.com/startup-insider-edge/i-built-a-desktop-app-to-fix-the-messiest-part-of-my-git-workflow-c432d0bc90f3?source=rss-a77250ac42f4------2) Mar 02, 2026- [The Anti-Inflation RPG That’s Actually Paying Out (And Integrating Pi)](https://medium.com/@zintaen/the-anti-inflation-rpg-thats-actually-paying-out-and-integrating-pi-2cfeda7191d1?source=rss-a77250ac42f4------2) Jan 21, 2026- [The “God Mode” Toggle: Architecting Gemini’s Personal Context](https://ai.plainenglish.io/the-god-mode-toggle-architecting-geminis-personal-context-151625e912f0?source=rss-a77250ac42f4------2) Jan 12, 2026- [Stop Prompting, Start Architecting: How I Built a “Universal” Second Brain in Gemini](https://ai.plainenglish.io/stop-prompting-start-architecting-how-i-built-a-universal-second-brain-in-gemini-c4c49a5a2761?source=rss-a77250ac42f4------2) Jan 12, 2026If you're standing up agentic AI, debating a RAG stack, or trying to cut inference cost in half — grab a slot. First call's on me.
I'm most useful to teams that are:
- Standing up Agentic AI or RAG for the first time and need a production-grade blueprint
- Running inference workloads that are slow, expensive, or both
- Migrating monoliths to micro-frontends / event-driven microservices without breaking SLAs
- Shipping AI features under SOC 2, PII, or HITL constraints
If any of that sounds like your week — say hi.
⚡ Built with late-night espresso, strong opinions, and a healthy fear of unbounded context windows.




