Personal career site
A static site connecting a professional profile, bilingual résumé, engineering projects, and long-term topics.
My work： Content direction and design review, with AI-assisted implementation and validation.
Problem and constraints
The original site centered on studying LLM inference. It needed to explain a broader professional background while keeping projects and technical notes easy to explore.
The site stays static, uses Markdown, and preserves existing article addresses. Chinese and English have separate URLs, and reading does not require JavaScript. There is no database or content administration service.
My contribution
I set the direction for a career homepage and topic spaces, then reviewed the information architecture and visual design. AI assisted with implementation and checks. This case describes design decisions, content organization, and collaborative delivery rather than claiming every line was written independently.
Key decisions
- Keep one source of résumé facts with independently written Chinese and English text.
- Curate documented projects and distinguish work in progress from validated work.
- Associate topics with existing content rather than duplicate articles.
- Preserve notes, tags, and RSS URLs; label original-language content when no translation exists.
- Use Solarized-inspired semantic colors for readable pages and printing.
Validation approach
Static builds, internal link checks, draft isolation, and browser checks are used to validate delivery. Browser checks cover language, narrow layouts, keyboard operation, themes, and print styles. Live performance still needs validation in the actual hosting environment.
Limitations and next steps
The site presents confirmed capability summaries. Detailed employment dates, the formal résumé name, education, and publishable outcomes still need curation. Missing facts are omitted instead of replaced with invented employers or numbers.