nam@redsun:~$ whoamihuman — not a bot

Vũ Đức Nam

CTO & Head of Projects at Redsun Ecosys

I'm Nam — a software engineer of 10 years, building ERP, POS and AI tools for Vietnamese businesses. I use AI every day. Decisions, accountability and the promises we make to customers still belong to people.

— Nam

Thủ Đức, Ho Chi Minh City · UTC+7

火 · FIRE ELEMENTstatus: burning
fuel
  • ERP
  • POS
  • AI agents
  • Kafka
  • CRM
  • DAG engine
  • the team
2 rolesCTO and head of projects, at the same time
3 productsBOS, SIPOS and Webino, side by side
63 use casesAI-first, 10 domains, in a single spec
Fire never asks when it gets to rest. It only asks what is left to burn.

01 — Who I am

A craftsman at heart.

At Redsun, I design architecture, plan products, and still sit down with developers to untangle problems one by one.

Outside work

  1. A dad who codes for his kid — a Grade 5 math practice app and a Socratic AI tutor that asks questions instead of handing out answers.
  2. Small tools for real jobs, like a lookup app for administrative documents — accurate down to the last word.
content/en.ymlEdit on GitHub
# a human, not a bot
element: fire
stack:
  - NestJS, Go, Rust
  - PostgreSQL, Redis
  - Kafka, BullMQ
  - Mobile apps
  - Claude API, Claude Code, Antigravity
focus:
  - AI agents for ERP
  - Event-driven architecture
  - Enterprise-grade products

02 — What I do

From architecture diagrams to products that run.

/architecture

Large-scale system architecture

Microservices, event-driven, database-per-service — for a multi-industry, multi-brand ERP platform.

NestJS · Go · Rust · Mobile · PostgreSQL · Kafka · BullMQ · Redis

/ai

AI in real products

An AI Gateway that calls business operations through function calling, agents that replace ERP screens, AI summaries and next-step suggestions in CRM.

Claude API · MCP · pgvector

/automation

Automation & omnichannel

A DAG-based workflow engine, customer-care campaigns, and marketplace integrations.

Zalo OA/ZNS · SMS · Email · TikTok Shop · Shopee

/leadership

Leading product & team

Product planning, writing specs, and coordinating a project team across PO, frontend and full-stack.

Product planning · Specs · Code review

03 — How I think

How I make decisions.

  1. AI proposes, people confirm.

    Agents can draft orders, build reports and suggest next steps — but anything that touches real data goes through a preview and a human confirmation.

  2. Conversation over screens.

    Business users shouldn't have to memorise ten screens. They say what they need; the system finds the right operation.

  3. Clean data first, AI second.

    Cleaning and standardising data is the foundation. Without it, any model just gives wrong answers faster.

  4. Architecture sized to reality.

    MVP first, scale later. Running cost is a design constraint, not an afterthought.

  5. Don't rebuild what already exists.

    Orchestrate existing services, integrate good open source. Save the team's energy for what only we can build.

04 — Contact

Message me directly.

About ERP, POS, AI agents for business — or just to talk shop. The person reading and replying is me, not a chatbot.

Thanks for stopping by. — Nam