Zaain Sabahat/ projects Back to terminal
§ F4 — Projects

The work, not the claim.

Three projects I've built and use day-to-day. One full equity research note, one quarterly screener, and one weekly macro framework. Real artifacts, not placeholders. Each one earned its place in my workflow before it earned its place on this site.

A full research note on NVIDIA.

NVIDIA is one of the few names where the bull case and the bear case both have credible fundamentals — that's exactly what makes it worth modelling carefully rather than reciting a thesis from memory.

The note builds a standalone DCF with explicit revenue and margin assumptions out to FY27, triangulates against peer multiples, and then runs three scenario paths (bear, base, bull) to bound the price target. The conclusion: BUY at $222, with the DCF mid-case at $235 — call it ~18% upside from the print date.

The work that matters here isn't the answer, it's the method. WACC is decomposed and defended; terminal growth is a real assumption, not 3% by reflex; sensitivity tables show where the valuation breaks. A recruiter reading this should be able to challenge any single input and see exactly where it flows through.

SubjectNVIDIA Corp.
TickerNVDA
SectorTech · Semis
RatingBUY
Price target$222
DCF value$235
WACC10.5%
Terminal g3.0%
Implied upside~+18%
FormatPDF · Excel
NVDA NVIDIA Corporation
EQUITY · NDQ
Rating
BUY
12-mo PT
$222
↑ 18% upside
DCF Value
$235
WACC
10.5%
Terminal g
3.0%
FY27e Rev
$210B
FY27e EPS
$6.10
Implied P/E
36.4×

// What the note covers

  • Bottom-up DCF with explicit FY24–FY27 revenue and margin paths, decomposed by segment
  • Peer multiple analysis against AMD, AVGO, INTC, plus broader hyperscaler comp set
  • Three-scenario range: bear ($165) · base ($222) · bull ($280) with explicit assumption deltas
  • Sensitivity tables on WACC × terminal growth and revenue CAGR × terminal margin
  • Risk register: customer concentration, geopolitical, valuation, technology cycle
Download full PDF report + 1 more report in catalogue

A multi-factor screener for the PSX universe.

A quarterly model that pulls in EPS history, sector P/Es, current price, and analyst-estimated next-quarter EPS for the full PSX universe — KSE-100, KMI-30, and broader all-share names. Roughly 100+ tickers refreshed every reporting season.

The output is a shortlist. Every name is scored against three rules: EPS Q/Q expansion, valuation within ±1.5× of sector P/E, and implied upside greater than 25%. Names that pass all three become candidates for the swing book; everything else gets monitored for the next quarter.

This isn't a black box — every formula is visible, every input is sourced, and every shortlisted name has to survive a manual fundamental review before it gets traded. The screener is the funnel; the work happens after.

UniversePSX · 100+ tickers
RefreshQuarterly · post-results
Factors3 · multiplicative
Rule 1EPS Q/Q ↑
Rule 2P/E vs sector ±1.5
Rule 3Upside > 25%
OutputShortlist · ranked
ToolingExcel · sparklines
Used since2024

// Methodology

  • Step 1. Pull EPS for the last 7 quarters per ticker; compute trailing-12-month EPS and Q/Q growth %
  • Step 2. Calculate P/E using current price ÷ T12M EPS; benchmark against sector P/E (SCStrade reference)
  • Step 3. Compute next-quarter expected EPS; project annualised forward EPS; build expected price under sector P/E
  • Step 4. Rank names by upside %; flag those satisfying all three rules
  • Step 5. Manual fundamental review on the shortlist — sector context, accounting quality, results day reaction

A weekly release calendar, mapped to cross-asset.

Every month, the US data cycle prints in roughly the same shape: ISM in week 1, CPI in week 2, FOMC in week 3, GDP and PCE in week 4. Knowing this isn't impressive on its own — but reading which release matters most for which trade is.

The framework is a workbook with 12+ tabs, one per primary release. Each tab logs: actual vs forecast, the surprise direction, and the historical reaction in DXY, gold, indices, and risk. Over time it becomes a database of how the market has actually responded — not how textbook says it should.

I read it on Sunday nights. Whatever's coming that week tells me which positions to size up, which to flatten, and which to skip entirely. The framework doesn't predict; it contextualises.

CycleUS · monthly
Releases tracked12+
PrimaryISM · NFP · CPI · GDP · PCE
SecondaryPPI · Retail · Housing · Claims
FedFOMC · Minutes · Speeches
Cross-assetUSD · gold · indices · risk
RefreshWeekly · Sunday review
ToolingExcel · 12+ tabs

// What's in the workbook

  • Master calendar — every release placed in its typical week and weekday, primary/secondary tagged
  • Per-release tabs — ISM Manufacturing, ISM Services, NFP, CPI m/m, CPI y/y, PPI, Retail Sales, Industrial Production, GDP advance, Core PCE
  • Surprise tracking — actual vs consensus, beat/miss flag, magnitude in standard deviations
  • Reaction logging — how DXY, XAU/USD, ES futures, and VIX moved in the 30-min and 4-hour window post-release
  • Fed speech tracker — Powell, vice chair, regional presidents, with hawkish/dovish flags
  • Jackson Hole archive — annual symposium notes