Platform Engineer · Cloud-Native Architect · Infrastructure Builder
The system is the core; I am its architect.
I design and operate production infrastructure where explicit boundaries, deterministic behavior, and controlled complexity matter.
My work sits at the intersection of Kubernetes, cloud infrastructure, distributed systems, observability, and automation — turning fragmented infrastructure into reproducible, operable systems.
Self-consistency
+
Explicit boundaries
+
Controlled complexity
↓
Higher-tier correctness
I tend to approach infrastructure as a system rather than a collection of services.
Complexity
│
├── eliminate what is unnecessary
├── isolate what must coexist
├── automate what can be declared
└── observe what cannot be assumed
│
▼
Predictable Systems
I care particularly about:
- Structural simplicity over accidental complexity
- Declarative infrastructure over manual operations
- Strong boundaries between tenants, environments, and failure domains
- Deterministic deployments and reproducible infrastructure
- Observable state instead of implicit assumptions
- Safe migrations without unnecessary downtime
- Automation that reduces operational entropy
Cloud
├── AWS
│ ├── EKS
│ ├── VPC / TGW
│ ├── IAM
│ ├── ALB / Gateway
│ ├── S3
│ └── CloudWatch
│
└── Aliyun
├── ACK
├── VPC / CEN
└── Cloud Infrastructure
My primary focus is building multi-region, multi-tenant cloud platforms with Kubernetes at the center.
Core technologies:
Kubernetes EKS ACK Terraform Helm GitOps AWS Aliyun
I have worked extensively with stateful middleware and production migrations:
Kafka
├── Cluster Operations
├── Replication
├── MirrorMaker
└── Live Migration
Zookeeper
└── Stateful Cluster Operations
RabbitMQ
├── Cluster Migration
└── Zero-Downtime Cutover
One of my major infrastructure projects involved reconstructing and migrating mission-critical Kafka, Zookeeper, and RabbitMQ infrastructure across Southeast Asian regions, with a focus on:
- Zero-downtime migration
- Cross-region network topology
- Client transparency
- Message ordering
- Data integrity
- Controlled cutover and rollback
The interesting part was never simply moving the servers.
It was preserving the system's behavior while its underlying topology changed.
I treat observability as a first-class part of infrastructure design.
┌──────────────┐
│ Runtime │
└──────┬───────┘
│
┌─────────────┼─────────────┐
▼ ▼ ▼
Metrics Logs Events
│ │ │
▼ ▼ ▼
Prometheus ELK Splunk
│
▼
Grafana
│
▼
System Insight
Stack:
Prometheus Grafana ELK Splunk node_exporter
I am particularly interested in the gap between:
"the system is running"
and
"we actually understand what the system is doing."
Kubernetes observability & Prometheus dashboard engineering
A collection of production-oriented Grafana dashboards and Prometheus queries for Kubernetes environments.
The goal is straightforward:
Make cluster state visible enough that operational decisions can be made from evidence rather than intuition.
Focus: Kubernetes · Prometheus · Grafana · Metrics · Production Observability
Declarative wallpaper delivery pipeline for Apple platforms
An exploration of engineering aesthetics through a fully automated content pipeline.
Git Repository
│
▼
Asset Validation
│
▼
Optimization
│
▼
GitHub Actions
│
▼
Edge Distribution
│
▼
SwiftUI Client
Focus: SwiftUI · GitHub Actions · Automation · Content Delivery
A GitHub-native image repository designed as a structured, version-controlled asset store.
Instead of treating wallpapers as static files, the project explores the idea of content as an immutable, automatable data source.
Native macOS audio player
A minimalist native macOS music player focused on predictable lifecycle management, constrained abstractions, and a clean interaction model.
Focus: Swift · macOS · Audio · Native Systems
My preferred infrastructure model looks roughly like this:
Git
│
▼
Declarative Configuration
│
┌───────────┴───────────┐
▼ ▼
Terraform Helm
│ │
▼ ▼
Cloud Layer Kubernetes Layer
│ │
└───────────┬───────────┘
▼
Production
│
┌───────┴───────┐
▼ ▼
Observability Automation
│ │
└───────┬───────┘
▼
Controlled State
The objective is not to eliminate complexity.
It is to make complexity explicit, bounded, and controllable.
Cloud
AWS · Aliyun
Container & Orchestration
Kubernetes · EKS · ACK · Helm
Infrastructure as Code
Terraform · GitOps
Networking
VPC · TGW · CEN · ALB · Gateway API
Middleware
Kafka · Zookeeper · RabbitMQ
Observability
Prometheus · Grafana · ELK · Splunk
Automation
GitHub Actions · Jenkins · GitLab CI
Systems
Linux · macOS
Languages
Python · Go · Bash · Swift · C/C++ · Java · SQL
I enjoy engineering systems at different layers — from cloud control planes and distributed middleware down to native applications.
There is a common thread between them:
reduce unnecessary complexity, define the boundaries, and make the system behave predictably.
Infrastructure is where I spend most of my time.
But the underlying interest is broader:
How do we build systems that remain understandable as they become larger?
Static Wiki therealmileslee.github.io
Professional Graph LinkedIn
Secure Gateway
trdli@ucdavis.edu
┌─────────────────────────────────────────────────────┐
│ │
│ Build systems. │
│ Reduce entropy. │
│ Make complexity explicit. │
│ │
│ — Miles Li │
│ │
└─────────────────────────────────────────────────────┘