Senior software engineer

Seven years of distributed systems.

I build cloud-native platforms for financial workflows — the kind where reliability isn't a feature, it's the cost of admission. Currently leading architecture for IRIS at Dragon Capital.

I work mostly in C# and Rust, deploy on Azure and Kubernetes, and care about the gap between what's shipped and what actually runs well in production. Based in Ho Chi Minh City.

Seven years of distributed systems.
Practice
Cloud-native · Distributed systems
Experience
7+ years · 2017 — present
Current role
Senior Software Engineer · Dragon Capital

01 · Engineering

Web App & Microservices

Cloud-native · C#, Rust, TypeScript · 2017 — present

I build web applications and microservice systems that need to keep running when something fails — distributed services, financial data pipelines, and order-management workflows on Azure AKS and Kubernetes.

Languages and tools I reach for: C#, Rust, TypeScript, Azure Service Bus, PostgreSQL, Redis, ElasticSearch, Terraform. The shape of the system matters more than the framework.

02 · Operations

Reliability & ops

Production reliability · CI/CD · on-call

I treat reliability as a feature, not a footnote. At Dragon Capital I modernized the IRIS CI/CD pipeline, improved database architecture, and rolled out observability that catches problems before traders do.

Deploys should be boring. If a deployment is exciting, the system is telling you something.

03 · Leadership

Mentor & lead

Tech lead · 12–15 engineers · AI-assisted workflow

I've led teams of 12–15 engineers at Reso, owned knowledge transfer and internalization of mission-critical systems at Dragon Capital, and mentored engineers through refactors and onboarding.

Recently I rolled out GitHub Copilot Enterprise across the IRIS team — onboarding, refactoring, and code comprehension got noticeably faster.

Career timeline

Seven years, five chapters.

Each chapter is a lesson — different teams, different problems, one through-line: distributed systems that actually work in production.

  1. 2017

    First production code at WiSky

    WiSky · Junior Software Developer

    Started as a junior building WiFi marketing platforms, attendance systems, and an internal web builder framework. Learned the value of caching, query optimization, and shipping.

  2. 2019 — 2022

    High-volume financial microservices with Amex OneAP

    Easy IT Solution · Amex OneAP · Middle Software Engineer

    Full-time on the Amex OneAP team building enterprise financial microservices. Shipped the CPG Gateway handling high-volume transaction requests, the Identity Security Service for auth, and optimized background workers for payment processing.

  3. 2022 — 2024

    Payment gateway and Azure-native platform at eTreem

    eTreem · Senior Software Engineer (contractor) · Azure AKS · Payrix

    Built the Payrix Gateway Service integrating enterprise payment infrastructure, designed a centralized Identity Server for the microservice ecosystem, and architected the platform on Azure Kubernetes Service with distributed observability via Elastic Stack and Jaeger.

  4. Now

    Leading IRIS architecture at Dragon Capital

    Dragon Capital · Senior Software Engineer · IRIS · OMS · Algo X · ISBD

    Own architecture and operational stability of IRIS — the core investment platform — plus OMS, Portfolio Rebalancing, ISBD data pipelines, and Algo X trading integration. Real money, real reliability constraints.

  5. Recent

    AI-assisted engineering workflow

    GitHub Copilot Enterprise · onboarding · refactoring

    Introduced an AI-assisted development workflow across the IRIS team. Accelerated onboarding, code comprehension, and refactoring — and changed how I think about leverage in engineering work.

Career domains

Three industries, seven years.

From WiFi marketing tools to fund management platforms handling real money — same engineering discipline, very different problem shapes.

01 · Retail & operations

Wifi Market Event · POS · CMS · CRM

WiSky · Reso · ExE · 2017 — 2019

At WiSky I built WiFi marketing platforms, attendance systems, and an internal web builder framework that accelerated site delivery. At ExE I integrated Azure Face Recognition into attendance tracking and built logistics support utilities for warehouse operations.

At Reso I led the development of CRM and POS systems for retail and F&B businesses — the operational backbone for hundreds of merchants — while managing a team of 12–15 engineers.

02 · B2B payment gateway

B2B Payment Gateway

Easy IT · eTreem · 2019 — 2024

At Easy IT Solution on the Amex OneAP project I shipped the CPG Gateway handling high-volume financial transaction requests, the Identity Security Service for auth, and optimized background workers for payment processing — enterprise-grade B2B payment infrastructure under Amex's engineering standards.

At eTreem I built the Payrix Gateway Service integrating enterprise payment infrastructure, designed a centralized Identity Server for the microservice ecosystem, and architected the platform on Azure Kubernetes Service with distributed observability via Elastic Stack and Jaeger.

03 · Financial · fund management

Financial & fund management

Dragon Capital · 2024 — present

At Dragon Capital I own architecture and operational stability of IRIS — the core investment platform covering portfolio management, OMS, Portfolio Rebalancing, ISBD data pipelines, and Algo X trading integration. Real money, real reliability constraints.

The IRIS platform supports research, portfolio management, front-office trading, middle office, reporting, and data distribution — the full lifecycle of fund operations.

Money doesn't lie. Production either works or it doesn't — everything else is a story we tell ourselves.

Personal note, after seven years operating financial systems in production.

Practice

Small habits.

  • 01 · Read the system before changing it

    Understand before editing. The fastest way to break a distributed system is to change it before you understand why it's shaped that way.

  • 02 · Make deploys boring

    If a deployment is exciting, something is wrong. Boring deploys are the result of good CI/CD, automation, and runbooks — not luck.

  • 03 · Migrate vendor code, eventually

    Vendor code is a starting point, not a permanent dependency. Plan the internalization early, even if it happens late.

  • 04 · Use AI as a pair, not a crutch

    Engineers who use AI will replace those who don't. But the value comes from your judgment, not the autocomplete.

Get in touch

Open to conversations.

I respond to thoughtful messages about cloud-native platforms, FinTech architecture, distributed systems, and engineering team modernization. Replies usually within a couple of days.