Nguyen Quoc Doan
Senior DevOps Engineer / Technical Architect
AWS · Azure · GCP
Kubernetes · GitOps
Kafka · CDC · NiFi
AI Agents · MCP
About
Platform Engineer and Technical Architect with 8+ years designing and shipping production systems
end-to-end — cloud infrastructure, backends, third-party integrations, and AI-enabled platforms.
Experienced across hybrid cloud-edge (Kubernetes, Kafka, CDC), Node.js microservices, and modern
integration patterns (MCP servers, GraphQL, verifiable credentials). Currently building regulated
identity & compliance flows (KYC/KYB, VO) on an AI workspace platform. Equally comfortable
architecting, coding, and owning delivery — CI/CD, observability, and 99.9%+ uptime at scale.
Core Expertise
Cloud & DevOps
5+ years
AWS (EKS, EC2, S3, RDS, MSK), Docker, K8s (EKS, K3s, K0s), Helm, ArgoCD, Terraform, Ansible, GitHub Actions, Bitbucket, GitLab CI, Prometheus, Grafana, Loki, ELK
Backend & Architecture
8+ years
Node.js, NestJS, Express, TypeScript, Microservices, GraphQL, REST, Event-driven systems
Data Engineering
4+ years
Apache Kafka, Kafka Connect, Debezium, Apache NiFi, Redis, PostgreSQL, ETL, CDC
Frontend & Integration
5+ years
React, Next.js, Vue.js, Micro-frontends
AI & Emerging Tech
Current focus
Multi-LLM routing (OpenAI, Gemini, Claude, LiteLLM), MCP connectors, AI agent workflows, production AI SaaS (7+ channels)
Experience
Phase 1 of 4
Jun 2018 – Oct 2020
Foundation in full-stack development and early DevOps practices. Built freelancer platform
and internal systems before transitioning into architecture and platform engineering.
LTV Coffee 🏆 Top 15 Startups Vietnam 2019
- Designed and developed freelancer platform using Node.js, Vue.js, and MoleculerJS.
Stack: Node.js, Vue.js, MoleculerJS
Phase 2 of 4
Senior Engineer | DevOps Engineer
Oct 2020 – Mar 2023
Scaled enterprise SaaS on AWS, automated CI/CD, and built integration platforms linking
DevOps workflows with business operations (SAP, approval hubs).
- Designed and deployed AWS infrastructure using Amplify, ECS, and Lambda for rapid scaling.
- Developed Next.js + Node.js microservices for internal management and integration platforms.
- Implemented Microsoft Entra Verified ID for secure identity across distributed microservices.
- Built workflow approval hub (Base.vn-style) integrated with SAP, linking DevOps and business teams.
- Automated CI/CD with Bitbucket Pipelines — 50% faster deployments, stable rollouts.
Stack: AWS (Amplify, ECS, Lambda), Node.js, Next.js, Microsoft Entra Verified ID, SAP, Bitbucket Pipelines
Phase 3 of 4
Mar 2023 – Mar 2026
Led architecture for hybrid cloud-edge systems: 120+ vessels, real-time CDC sync, GitOps on
EKS/K3s/K0s, and AI SaaS platform at 99.95% uptime. Onsite leadership in Korea (Hinavi).
Hinavi Project Onsite Korea
- Hybrid cloud–edge: 120+ vessels connected to AWS shore via Starlink.
- Real-time CDC with Kafka Connect & Debezium between shore and vessel PostgreSQL.
- Shore on EKS, vessels on K3s — autonomous operations at sea.
- Istio service mesh for routing and fault tolerance.
- GitOps (ArgoCD, Helm, Kustomize), zero-downtime rollouts.
- Batch optimization cut RDS + Kafka usage by 60%.
- Full Prometheus + Grafana + Slack alerting.
Stack: AWS EKS, MSK, RDS, Kafka, Debezium, K3s, ArgoCD, Helm, Istio, NestJS, React
Oasis Project 🏆 Top Individual Excellence 2024
- Hybrid platform: AWS EKS (shore) + K0s (edge IoT).
- Terraform IaC: Lambda, SNS/SQS, Kubernetes clusters.
- Apache NiFi + SQS for cross-DB and IoT event sync.
- GitHub Actions, Helm, AWS Secrets Manager.
Stack: AWS EKS, Lambda, SNS, SQS, Terraform, NiFi, K0s, PostgreSQL, Prometheus, Grafana
Hilios — AI Agent Platform
- Core DevOps for AI SaaS chatbot on 7+ platforms (Zalo, Messenger, WhatsApp, Slack…).
- Multi-LLM routing (OpenAI, Gemini, Claude): 99.95% uptime, 73% latency reduction.
- GCP/GKE cost -36%; re-architected with Node.js + Bun.
Phase 4 of 4
KYC/KYB · Verifiable Credentials · MCP · AI Agent Integrations
Apr 2026 – Present
Production integrations on myHub (mySMB AI workspace) for Australia's AML/CTF Tranche 2.
VerifyMe handles KYC/KYB; VO issues reusable verifiable credentials; MCP servers expose both
to Claude agents as callable tools inside the workspace.
- MCP server development — Node.js/Express on Azure Container Apps; tools for KYC/KYB invite, async risk webhooks, VO VC issuance & presentation (GraphQL).
- Marketplace connectors — myHub plugins packaging MCP endpoints, widgets, and agent skills for the extensibility platform.
- KYC/KYB → VC pipeline — VerifyMe 2-phase verification feeds VO credential lifecycle: issue to wallet, present, reuse, expire.
- AI agent workflows — agents orchestrate compliance vetting end-to-end via MCP without bespoke UI per vendor.
Stack: Node.js, TypeScript, Express, MCP, VerifyMe API, VO (GraphQL), Verifiable Credentials, Azure Container Apps
Additional Experience
Cyfeer — Contract DevOps / Platform Engineer
Thadico · Sun Group · Cyhome
- Thadico: Production K3s, HAProxy, Traefik, Prometheus, Grafana, Loki, ELK, MinIO backups.
- Sun Group: Azure AKS, GitLab CI, ACR, ArgoCD, Argo Rollouts, Terraform, Key Vault.
- Cyhome: Docker Compose → AWS EKS migration (9 nodes), zero-downtime blue-green, 40% cost cut.
Dou Networks Co., Ltd — Software Engineer
Jan 2015 – Jun 2018
- Enterprise platforms with Node.js, Java, AngularJS, PostgreSQL.
- Introduced Docker and Jenkins CI/CD — first DevOps exposure.
Key Achievements
🏆 Top Individual Excellence 2024 Architecture & system reliability leadership
🏆 Top 15 Startups Vietnam 2019 LTV Coffee
💰 Cost savings 36% GCP (AI); 40% AWS optimization
⚡ Performance 73% latency cut (3s → 800ms); 75% API speedup
✅ Reliability 99.95% uptime (1M+ msgs/day); 99.9% for 120+ edge devices
👥 Leadership DevOps teams in hybrid cloud-edge environments
Education & Languages
FPT University
Advanced Diploma in Software Engineering (2012–2015)
Fast-track program: OOP design, user-oriented interface design.
Languages
English — Intermediate