Vòng tròn Năng lực

Draw the actual boundary, not the flattering one

Nine years in the ER gives you a real circle: triage, resuscitation, medication administration, family communication during acute events. Prognosis is outside it — that's a physician's call, and even they hedge. Your mistake wasn't ignorance; it was treating a confident tone as evidence of competence.

Do the boundary audit

Write down three columns: I can defend with data (e.g., "I can predict deterioration risk in the next hour based on vitals and labs"), I can describe but not predict (e.g., "I know sepsis protocols, but not long-term survival curves"), I only assume (e.g., "recovery odds for this comorbidity profile"). If you can't cite a specific study, guideline, or hard number, it goes in column three.

Add the Munger test

Before you speak again, ask: Can I argue the pessimistic case better than the attending can? If not, you're not qualified to hold the opinion — say "I don't know" out loud. That's not weakness; it's the boundary holding. And note: your real skill — calming a family without making a prognosis — is inside your circle. Bet boldly there.

$50k MRR is real, but that's not the question. The question is: why do you think you understand fintech? Customers requesting a feature is evidence of demand, not evidence of your competence to build it safely.

Run the Buffett test: can you articulate, in writing, the regulatory landscape, the fraud vectors, the counterparty risk, and the failure modes of the payment rails you'd touch? If you can't explain why a chargeback cascade or a KYC failure would break your model — and what you'd do when it does — you're outside the circle. That's not shameful; it's the boundary.

The discipline here isn't "don't build it." It's "don't bet three months on your own ignorance." So do this: spend one week, not three months, interviewing two operators who've shipped fintech features. Ask them what they know now that they wish they'd known. If their answers sound like a foreign language, you have your answer — partner with someone inside the circle, or shelve it.

Chasing shiny revenue is exactly how founders blow up good SaaS businesses. The Dunning-Kruger trap is thinking product instincts transfer to regulated money. They don't.

Your two-year tech streak isn't proof of edge—it's a sample size too small to separate skill from a bull market tailwind. The biotech loss is the gift: it drew your circle's boundary in blood. Diversifying into sectors you don't understand isn't diversification; it's multiplying ignorance. Buffett's discipline isn't about avoiding risk—it's about knowing which risks you can price. Ask yourself the Munger test: can you argue the bear case for your tech picks better than a short-seller can? If not, you don't hold an opinion, you hold a hope. Your real edge ends where your ability to explain why a stock will move—not just that it might—breaks down. Map it: list every position you hold, and for each, write the three variables that would change your thesis. If you can't name them, you're outside the circle. Double down on tech only where you pass that test. For biotech, either study it for six months until you can articulate the regulatory and clinical pathways, or permanently abstain. The discipline isn't "never touch"—it's "don't bet until you understand." Your edge is narrow and deep, not wide and shallow. Respect that, and the market stops being your teacher and starts being your scoreboard.