CoreWeave, Inc. -- 4.0/10 -- $82.24
Gate result: Three NOs. Raw composite score of 5.3 capped at 4.0 maximum. Flagged as Speculative / Does Not Meet Quality Bar. CoreWeave is a leveraged infrastructure intermediary with no competitive moat, no free cash flow, and no meaningful public company track record. The extraordinary revenue growth and $66.8B backlog are real, but the business model -- borrow heavily to buy GPUs, lease to AI companies -- has no margin of safety in a downturn. Not investable under the quality framework regardless of the AI infrastructure opportunity.
| Dimension | Score | Weight | Weighted |
|---|---|---|---|
| Financial Trends | 5.0 | 25% | 1.25 |
| Thematic Exposure | 7.5 | 25% | 1.88 |
| Management Quality | 4.0 | 20% | 0.80 |
| Investor Sentiment | 5.5 | 15% | 0.83 |
| Concerns, Catalysts & Risks | 3.5 | 15% | 0.53 |
| Raw Composite | 100% | 5.3 | |
| Capped (3 gate failures) | 4.0 |
CoreWeave is a GPU cloud infrastructure provider headquartered in Roseland, New Jersey, offering NVIDIA-powered compute as-a-service to hyperscalers, AI labs, and enterprises. The company was founded in 2017 as a cryptocurrency mining operation and pivoted to AI cloud infrastructure. It IPO'd on NASDAQ in March 2025 at ~$40/share. FY2025 revenue was $5.1B, up 168% YoY. Calendar fiscal year (December FYE).
The investment case is a bet on sustained AI infrastructure demand at extreme leverage. The bull case centers on world-class revenue growth ($5.1B to guided $12-13B in FY2026), a $66.8B contracted backlog with blue-chip customers (Microsoft, Meta, OpenAI), first-mover NVIDIA partnership status (Exemplar Cloud for GB200), and a path to 70% EBITDA margins on mature contracts. CoreWeave is building the picks-and-shovels infrastructure for the AI revolution -- if AI spending scales as expected, the demand runway extends through 2030+.
The bear case is that this is a leveraged intermediary with no moat, priced for perfection in a market that punishes execution misses. D/E of 8.94x with $14.6B+ in debt, interest expense consuming 24% of revenue, a $30-35B capex plan for FY2026 that requires continuous capital markets access, and GAAP losses of $1.2B. The business model -- borrow to buy NVIDIA GPUs, lease to AI companies -- works spectacularly in an upcycle but has zero margin of safety if AI spending decelerates, hyperscalers insource, or capital markets tighten. Customer concentration on a few hyperscalers, NVIDIA hardware dependency, and a CEO whose background is in commodities trading (not enterprise tech) add to the risk profile. The stock is down 56% from its $187 high with 19% short interest -- the market is already pricing significant skepticism.
| Price | $82.24 | FY2025 Revenue | $5.1B (+168% YoY) |
| Market Cap | ~$32B | Adj. EBITDA Margin (Q4) | 57% |
| Contracted Backlog | $66.8B | GAAP Net Loss (FY25) | -$1.2B (-23.3%) |
| Debt / Equity | 8.94x | Interest Expense (FY25) | $1.22B (~24% of rev) |
| 52-Week Range | $33.51 - $187.00 | FY2026 CapEx Guide | $30-35B |
| Insider Ownership | 39.12% | Current Ratio | 0.46 |
CRWV receives a raw composite score of 5.3/10, capped at 4.0/10 due to all three quality gates failing. Financial Trends (5.0) reflects world-class revenue growth offset by extreme leverage, negative GAAP profitability, and massively negative FCF. Thematic Exposure (7.5) is near-maximal -- CoreWeave is at the epicenter of the AI infrastructure buildout. Management Quality (4.0) reflects impressive execution on growth but an unproven team with no public company track record through adversity. Investor Sentiment (5.5) shows deep divergence between bullish analysts and bearish short sellers. Concerns and Risks (3.5) reflects the severe, multidimensional risk profile.
Bull case (~$120-150, +50-80%): AI spending accelerates, CoreWeave executes on $12-13B FY2026 revenue guide, adj. EBITDA margins expand toward 70% on seasoned contracts, new customer wins diversify the base, NVIDIA partnership deepens with Rubin/Vera hardware. The $66.8B backlog converts into visible, recurring revenue. Stock re-rates toward analyst consensus targets as execution proof points accumulate.
Base case (~$70-90, -15% to +10%): Revenue grows strongly but misses the high end of guidance. Capital markets remain accessible but cost of debt increases. Customer concentration improves modestly. Stock range-bound as bulls and bears debate whether the leveraged model is sustainable. Short interest remains elevated.
Bear case (~$30-50, -40-60%): AI spending cycle slows or hyperscalers insource more aggressively. Capital markets tighten, making the $30-35B capex plan harder to fund. Customer concentration risk materializes (contract delays, renegotiations). The leveraged model amplifies downside -- fixed debt payments against potentially variable revenue. Stock retests the 52-week low of $33.51.
Bottom line: CoreWeave is a high-growth, extremely leveraged infrastructure intermediary that is the purest play on AI compute demand. The growth numbers are extraordinary, but the risk profile is among the most severe in the public markets. Three quality gate failures cap the score at 4.0. NOT ACTIONABLE under the quality framework. Revisit only if (1) GAAP profitability is achieved on a sustained basis, (2) leverage declines meaningfully below 5x D/E, or (3) the company demonstrates resilience through a meaningful deceleration in AI spend.
Key catalysts and monitoring points:
- Q1 FY2026 earnings (~May 2026): First quarter toward the $12-13B annual revenue guide. Watch for revenue acceleration in absolute dollars, adj. EBITDA margin trajectory, and any update on the capex cadence within the $30-35B plan.
- Customer diversification: Enterprise customer wins beyond hyperscalers. New logos like CrowdStrike, Rakuten, and MercadoLibre are positive but still small relative to Microsoft/Meta/OpenAI concentration. Second and third anchor customers would materially de-risk the model.
- Capital markets access: CoreWeave needs continuous debt/equity issuance to fund $30-35B in capex. Monitor credit spreads, debt covenants, and any signs of financing stress. No debt maturities until 2029 is a mitigant, but new issuance costs matter.
- NVIDIA relationship and hardware transitions: Bringing Rubin, Vera, and BlueField to market in 2026. Execution on hardware transitions is a key differentiator vs. competitors. Any change in NVIDIA partnership status would be highly significant.
- AI spending trajectory: If hyperscaler capex guides moderate (watch MSFT, META, GOOGL earnings), downstream demand for CoreWeave could slow. The 5-year take-or-pay contracts provide buffer but new bookings could decelerate.
- Short interest evolution: 19% short float reflects significant institutional conviction against. Declining short interest would signal improving confidence; further increases would confirm growing skepticism.
- GAAP profitability path: Adj. EBITDA is positive and growing (57% margin in Q4), but GAAP losses persist due to interest expense and depreciation. Monitor the gap between adjusted and GAAP metrics -- it should narrow over time if the model is working.
For the full analysis, see the Business Model, Financials, Thematics, Management, Valuation, and Sentiment pages.