Turning business processes into reliable software.
把業務流程，變成可靠的軟體。
I’m Arthur Lin (林晉宇), a system analyst and full-stack engineer. I start from the interview and the spec, then build the API, the data, and an interface that can actually ship — including automation such as LINE.
Sample conversation, not usage data
Selected work
Work that is public and real. No sample projects.
LINE group assistant
LINE 群組助理
A LINE group assistant: AI Q&A with web search, chat summaries, reminders, expense splits, keyword replies, and a daily news push.
Next up: your project?
Start by writing down the scope.
Tools
Small tools live on a subdomain, for example tools.royalints.com, and are listed here.
No public tools yet
A card appears here once a subdomain is actually serving.
tools.royalints.comThe process, and the system that runs it
Arthur Lin
System Analyst & Full-Stack Engineer ・ Taipei
I do system analysis and full-stack engineering. A project starts by writing down who it’s for, where the data comes from, and what is out of scope — then the API, the data layer, and a version that can be deployed. Royalints is the name I use for freelance work.
System analysis
Interviews, flows, and API and data specs written so they can be built.
Full-stack
TypeScript services, the data layer, and the interface, through to deploy.
Automation
Things like the LINE group assistant: Q&A, reminders, scheduled pushes.
中文 and English
Specs, interfaces, and email in Chinese or English.
Services
One slice, or from the spec through to deploy. The scope is written down first.
System analysis
Interviews, flows, and boundaries, including what is out of scope this time.
- Interviews and flows
- API and data specs
- What done means
Full-stack development
TypeScript services, the data layer, and the interface, through to a deployable version.
- TypeScript / Node.js
- Interfaces and internal tools
- Deployment
Automation
Schedules, notifications, and third-party APIs. A LINE bot plus a cloud scheduler is one example.
- LINE group bots
- Q&A, reminders, pushes
- Cloud Scheduler
Existing systems
Read the current behavior and the risks, then change it in small steps. No schedule before the code has been read.
- Read what it does now
- Small changes
- Schedule after the code
Have something in mind?
Let’s build it.
Tell me the process, or where it hurts. Replies within 1–2 business days 把需求做成可以上線的系統。