Opportunity Engine
One amount. One stock. One decision — weighed by every layer NFAI can measure today, and honest about the ones still coming online.
NFAI's read · Taiwan Semi (TSM)
Confidence
LowPutting $10,000 into Taiwan Semi (TSM) buys about 24.74 shares — roughly 11.6% of the book. It reads balanced — nothing decisive for or against it today. NFAI can't yet see this name's intrinsic value or company-specific options, so its confidence is low.
Buys
24.74 sh
New book weight
11.6%
Marginal day-risk
+$217
Layers online
6 of 6
An orchestrated read — it adds no new forecast. Confidence is capped by how many layers are online, so what NFAI can’t yet see only lowers it. Each analyst below opens its full view.
A reasonable starter — buy about 24.74 shares (11.6%) small and add on confirmation.
The committee’s read
6 analysts weighing inIt contributes 14.7% of book risk for 11.6% of the capital — it pulls its weight in risk terms.
Trend tactics work: buy strength, give winners room, trail stops up.
Over ~3 months it is trailing SPY by −0.7% (67th percentile of the book).
Annualized volatility of 46.6% — average for this book (64th percentile).
At 11.6% of the book it's within the 12.0% single-name target — a clean starter size.
Shares are the expressible vehicle for this name today. A company-specific options read (chain and implied vol) is a separate analyst still coming online.
Still coming online
14 layers · the roadmapThe committee votes equally: of 6 seated members, none has yet cleared its sample bar to earn trust beyond the equal baseline. 14 subsystems produce views that aren't graded yet.
Intrinsic valuation
estimates fair value from fundamentals — needs a fundamentals feed (revenue, margins, cash flow); the /valuation deal engine covers explicit-input cases today
Options analytics
prices the best options expression for this name — needs a per-name options chain / implied-vol surface; only SPY has a live signal today
Sector strength
ranks the name's sector against the market — derivable from sector ETF series already fetched — a fast follow
Risk scenarios
prices what-if paths for this add — the /scenarios engine (lib/scenario.ts) exists; wiring it per-opportunity is a follow
Macro sensitivity
reads sensitivity to rates, credit and growth — a seam exists (realyield / credit betas); per-name attribution isn't wired to this surface yet
Earnings & event risk
flags earnings and events inside the horizon — needs an earnings-calendar feed
Analyst estimates
aggregates sell-side estimates and revisions — needs an estimates feed
Insider activity
reads insider buying and selling — needs a Form 4 feed
Institutional ownership
tracks 13F ownership changes — needs a 13F feed
Return distribution
shows the range of outcomes, not a point estimate — requires a calibrated return model; NFAI won't fabricate a distribution
Multi-idea optimizer
optimizes across several candidate ideas at once — the /capital optimizer (lib/optimizer.ts) exists; multi-idea intake is a future step
Bull & bear case
states the strongest bull and bear cases — reserved for a reasoning module grounded in evidence
AI thesis
writes the thesis in plain words — NFAI won't auto-write a thesis it can't ground in evidence
Recommendation learning
grades past recommendations and earns trust from outcomes — the recommendation ledger starts recording now; grading begins once outcomes mature
Key risks
Partial picture. Intrinsic value, earnings risk and company-specific options are still coming online, so this read is deliberately incomplete.
How confident, and why
NFAI stores this recommendation so it can later grade its own calls against what actually happens — the start of the learning loop. to add it to NFAI’s recommendation ledger.
Taiwan Semi (TSM) · price $404.25 · book value $187,285 · data live as of 2026-07-31. The universe is the 12 names NFAI continuously tracks, plus SPY. Portfolio impact is exact from the correlation block; momentum is derived from daily bars; valuation and company-specific options are not tracked per name yet. Educational tool, not investment advice.
