About Jeff
I'm Jeff. For more than 20 years my job has been international tax: serving multinational groups on cross-border investment structures, group reorganizations, transfer pricing, and in recent years the hottest topic in the field, the OECD Pillar Two / GloBE global minimum tax.
I have also invested in US stocks for years, holding twenty-plus individual names and ETFs. My habit when a 10-K lands differs from most investors: they go to EPS first, I turn to the income tax footnote. A one percentage point move in the effective tax rate often says more about earnings and valuation than whatever narrative the market is trading that week.
The third pillar is AI and automation. I pull SEC EDGAR and market data with Python, deep-read filings and regulations with AI tools, and run a multi-source verification protocol that keeps AI hallucination out of the output. With the research workflow automated, one person can produce close to team-level work.
Background
| Dimension | Detail |
|---|---|
| Practice | 20+ years in international tax, serving multinational groups |
| Academic | BA in Accounting + MS in Accounting + MS in Finance |
| Specialization | OECD Pillar Two / GloBE, cross-border tax structuring, M&A tax due diligence, transfer pricing |
| Investing | US individual stocks + ETF portfolio, long-term holdings |
| Technical | Python data automation (SEC EDGAR, yfinance), AI research workflow design |
| Languages | Traditional Chinese + English, bilingual |
What this site covers
- Deep dives into the tax structures of major US companies (where Apple's, Google's, and Microsoft's ETRs actually come from)
- Cross-border tax essentials for non-US investors in US stocks (estate tax, dividend withholding, ADR mechanics)
- How the Pillar Two global tax reform hits US tech stocks
- Hedge fund 13F tracking and analysis methods
- The complete Python + AI workflow for automating investment research
Writing principles: every number carries a source and a date, every case states its assumptions, and everything AI-assisted passes human review and multi-source fact-checking. Professional content is only worth reading if it can be trusted; that part is not negotiable.