Welcome to My Technical Blog!

Hi there! I'm Tommy Zhang, a Full-Stack Developer based in Miami. I spend most of my time these days building AI agents and the infrastructure that keeps them running in production. This blog is where I write down what I learn along the way.

What to Expect

Here's what actually shows up here, in rough order of how often I publish:

Weekly Digest

Every week I pull the GitHub Trending Top 5 from each day, deduplicate by repository, and rank what's left by how many days each project held its spot. The result is a digest of what the open-source world was actually paying attention to that week — not just a snapshot of one morning. This is the most consistent thing on the blog.

Infrastructure

Deep dives into systems I've built and run. The AI Gateway post is the template for this category: a self-hosted LiteLLM gateway on AWS EC2 with auto-failover, Redis caching, Prometheus/Grafana monitoring, and Cloudflare Zero Trust in front of it. Five Docker containers, seven security layers, about ten dollars a month. Real numbers, real tradeoffs, real things that broke.

Frontend

Patterns I've hit in production and wanted to write down properly — like the Sync Fetch pattern for escaping deeply nested async/await chains.

Algorithms

Notes from working through fundamentals. Less polished, more "here's how I finally got this to click."

My Background

I previously led digital transformation projects for five years across E-commerce, Banking, Electronic Payment, and Supply Chain before moving into full-time software development. That background is the reason I tend to ask "what does this cost and who actually uses it" before "what's the cleanest abstraction."

Since then the path has been:

• JFK Trading (2024–2025) — React Native warehouse app, Django REST APIs, Redis caching and Celery queues that cut response times by 40%.
• Clarium Managed Services (2025) — Next.js sites with Playwright end-to-end coverage, a SAP API integration, and an n8n pipeline that used OCR and an LLM to pull invoice data straight out of images.
• Aroma360 (2025–2026) — an AI shopping agent as a Shopify App, letting customers discover, compare, and buy through plain conversation. Also a metafields management tool with diff previews, snapshot rollback, and audit trails, because bulk-editing a live store with no undo is terrifying.
• Earthrobotics (2026–present) — leading AI integration. Designing and building AI agents and the agent skills infrastructure underneath them.

I studied Computer Science at St. Thomas University and Florida International University.

What I'm Working On Now

Agents. Specifically the unglamorous part of agents — how you package expertise so an agent can load it on demand, how you keep a fleet of them observable, and how you stop a model with tool access from doing something expensive and irreversible at 3am.

That's the thread running through most of what I'll be writing next.

Let's Connect

• GitHub: github.com/tommyy1708
• LinkedIn: linkedin.com/in/hua-zhang-fiu22
• Site: tommyz.blog

Final Thoughts

This blog is as much for me as it is for you. Writing something down forces me to find the parts I only thought I understood. If a post saves you an afternoon of debugging, even better.

Whether you're deep in this or just starting out, I hope you find something useful. Reach out anytime.

Happy coding! 🚀

---

Originally published January 11, 2025. Last updated August 16, 2026.