Alex Sacerdote founded Whale Rock Capital Management, a technology-focused investment firm. His investing framework looks for technology adoption inflections, durable competitive advantages and earnings power that the market may underestimate. The lessons below draw on his direct interviews and talks; dated investment views are presented as views, not settled outcomes. — Graham & Doddsville Interview.

Part 1: The Origins & The Amazon Epiphany
- The Fidelity training ground: Sacerdote credits Fidelity with giving him the time and colleagues to develop deep technology-sector research skills. — Graham & Doddsville Interview.
- Meeting Jeff Bezos: Early in his Fidelity career, he attended Amazon’s first investor day and had lunch with Jeff Bezos. — Graham & Doddsville Interview.
- Early skepticism: His Amazon recommendation met skepticism from colleagues who focused on the company’s losses and apparently high valuation. — Graham & Doddsville Interview.
- The 'buy the leader' thesis: He recalls recommending Amazon while shorting other retailers, an early example of backing the emerging leader in a platform shift. — Boston University Fireside Chat.
- Recognizing the internet shift: His experience at an internet startup helped him recognize how online commerce could reshape incumbent businesses. — Graham & Doddsville Interview.
- Ignoring high early valuations: He learned that an apparently high valuation can obscure a much larger adoption opportunity when a technology is still early. — Graham & Doddsville Interview.
- The value of specialization: After becoming a technology specialist at Fidelity, he focused on understanding the mechanics of technological disruption. — Graham & Doddsville Interview.
- Early career lessons: Investment banking, an internet startup and Fidelity gave him different perspectives on technology companies and their markets. — Graham & Doddsville Interview.
Part 2: The S-Curve Framework
- The flat start: New technologies can spend years in a slow phase while cost, complexity and missing ecosystems constrain adoption. — Graham & Doddsville Interview.
- The inflection point: When barriers fall and adoption broadens, unit growth can remain unusually strong for several years. — Graham & Doddsville Interview.
- Identifying the shift: His S-curve framework looks for the point where a technology moves from early adoption toward the mainstream. — Graham & Doddsville Interview.
- Predictability of demand: At an adoption inflection, a growing customer base can make demand more visible, although execution risk remains. — Graham & Doddsville Interview.
- The plateau phase: As penetration rises, growth eventually decelerates; he watches where a technology sits on that curve. — Graham & Doddsville Interview.
- Mapping historical cycles: He has applied the adoption-curve lens across successive technology cycles, including PCs, the internet, mobile and cloud. — Graham & Doddsville Interview.
- Avoiding the hype cycle: Distinguishing a promising technology’s slow early phase from a genuine adoption inflection is central to his process. — Graham & Doddsville Interview.
- The duration of exponential growth: An inflection can create a multi-year runway for unit growth that conventional forecasts may underestimate. — Graham & Doddsville Interview.
- Overlapping curves: Amazon illustrates how one company can benefit from more than one major adoption wave: online retail and then cloud infrastructure. — Graham & Doddsville Interview.
- Continuous application: Sacerdote describes the S-curve as a long-running core framework for Whale Rock’s technology investing. — Graham & Doddsville Interview.
Part 3: The AI "L-Curve" & Infrastructure
- The L-Curve: In June 2026, he characterized enterprise AI adoption as still under 1% and its potential rise as closer to an “L-curve” than a conventional S-curve. — Invest Like the Best Interview.
- Friction-less adoption: He contrasted slower, installation-heavy technologies with browser-accessible AI tools as a reason adoption could accelerate quickly. — Invest Like the Best Interview.
- The scale of AI penetration: He estimated in June 2026 that advanced agentic AI use among knowledge workers was about 10 basis points and forecast a rise toward 15% over four years; these were his estimates, not audited market measurements. — Invest Like the Best Interview.
- Hardware decommoditization: At the 2026 Sohn panel, he argued that AI infrastructure demand was making formerly commoditized hardware components more technically differentiated. — Sohn Conference Panel.
- The foundational model oligopoly: He viewed the leading frontier-model market in June 2026 as consolidating around OpenAI, Anthropic and Google, while acknowledging that the outcome remained uncertain. — Invest Like the Best Interview.
- Anthropic's trajectory: He described Anthropic as gaining users and revenue rapidly in 2026; this was his investment assessment, not an independently audited revenue figure. — Invest Like the Best Interview.
- The coding unlock: He singled out coding as an early breakout AI use case and estimated a large potential market for coding agents. — Invest Like the Best Interview.
- The golden age of hardware: He argued in 2026 that rapid token-demand growth could sustain demand for specialized AI hardware despite efficiency gains; his 14-fold versus two- to threefold figures were projections. — Sohn Conference Panel.
Part 4: Underappreciated Earnings Power
- Exponential vs. linear thinking: His framework expects earnings at a successful adoption inflection to rise nonlinearly rather than follow a smooth straight-line forecast. — Invest Like the Best Interview.
- Modeling the future: He models earnings several years ahead because a technology leader that looks expensive on current profits may be cheaper relative to its later earnings potential. — Invest Like the Best Interview.
- The Nvidia entry: Looking back, he argued that Nvidia’s AI opportunity made its future earnings power much larger than contemporaneous valuation measures implied. — Invest Like the Best Interview.
- The Tesla entry: He said Whale Rock’s Tesla thesis depended on much higher vehicle volumes and margins than consensus forecasts then assumed. — Invest Like the Best Interview.
- Getting businesses for free: He used Amazon as an example of a retail investment that later gained a second valuable growth engine in AWS. — Invest Like the Best Interview.
- Exiting the curve: He observed that US electric-vehicle adoption slowed around 10–15% penetration, illustrating how a presumed S-curve can stall. — Invest Like the Best Interview.
- Apple's early cycle: He recalled that Apple’s smartphone growth made its later earnings far larger than investors initially expected. — Graham & Doddsville Interview.
- The illusion of expense: A high multiple on current earnings can mask a technology leader’s potential earnings if adoption is near an inflection. — Graham & Doddsville Interview.
- Identifying trough valuations: Rather than rely only on trailing valuation, he estimates adoption speed and the earnings that could follow. — Graham & Doddsville Interview.
Part 5: Durable Competitive Advantage & Moats
- The necessity of moats: He looks for both technology-driven growth and a durable competitive advantage that lets the company retain value. — Graham & Doddsville Interview.
- Internet moats: For internet businesses, he identifies network effects as one possible source of durable advantage. — Graham & Doddsville Interview.
- Software standards: In software, an industry standard and the associated switching costs can reinforce a leader’s position. — Graham & Doddsville Interview.
- E-commerce scale: He cited Amazon’s retail scale and logistics as competitive strengths, while treating their durability as an investment judgment. — Graham & Doddsville Interview.
- Hardware monopolies: In hardware, critical intellectual property can help a specialist sustain a strong position. — Invest Like the Best Interview.
- Scale as an AI barrier: He described frontier AI labs as capital-intensive and treated their ability to raise financing as an important competitive factor. — Invest Like the Best Interview.
- Recursive AI improvement: He discussed coding outputs fed back into model development as one form of recursive improvement, particularly for coding-focused models. — Invest Like the Best Interview.
- The new Rule of 40: For chip investments, his modified “Rule of 40” adds AI-related business exposure to market share in the relevant category; it is a heuristic, not the conventional financial Rule of 40. — Invest Like the Best Interview.
Part 6: The "Whale Rock Learning Machine"
- Compounding knowledge: He describes Whale Rock as a learning organization that seeks to compound knowledge across technology cycles. — Graham & Doddsville Interview.
- Team dynamics: He emphasizes internal communication so that individual research becomes shared team insight. — Graham & Doddsville Interview.
- Meeting volume: In a 2026 interview, he said Whale Rock conducted roughly 2,500 face-to-face management meetings annually; the historical figure was lower in 2015. — Boston University Fireside Chat.
- The scuttlebutt approach: He draws on Philip Fisher’s scuttlebutt approach, speaking with companies and their customers, suppliers and competitors to test a thesis. — Invest Like the Best Interview.
- Private market context: He said a portion of Whale Rock’s management meetings involved private companies, giving context for emerging competitors. — Invest Like the Best Interview.
- Structural stability: He described a small, experienced investment team whose accumulated knowledge helps it study successive technology cycles. — Invest Like the Best Interview.
- Translating data to conviction: The purpose of primary research is to turn information into differentiated insight, not merely to maximize meeting volume. — Graham & Doddsville Interview.
- Intellectual curiosity: He linked intellectual curiosity and sustained critical thinking to his choice of a buyside investing career. — Graham & Doddsville Interview.
- Adapting the machine: He has discussed using AI to aid research while retaining direct company conversations and analyst judgment. — Invest Like the Best Interview.
Part 7: Volatility, Risk & The Long/Short Model
- Accepting volatility: He says volatility is inherent in growth-technology investing; long and short positions may dampen it but cannot remove it. — Boston University Fireside Chat.
- Market humility: His advice is to avoid overconfidence during strong performance and avoid losing perspective during drawdowns. — Boston University Fireside Chat.
- The TMT long/short advantage: He favors a long/short structure in technology because disruption produces both beneficiaries and vulnerable incumbents, while shorts can moderate some volatility. — Boston University Fireside Chat.
- Shorting criteria: On the short side, he looks for businesses whose competitive position is weakening and whose future earnings may be overestimated. — Graham & Doddsville Interview.
- The enterprise software short: At the 2026 Sohn panel, he argued that some CIO budgets were shifting toward AI tokens and away from established software, while noting that software remains sticky. — Sohn Conference Panel.
- Legacy AI products: He doubted that every incumbent software firm would produce a compelling AI application, but allowed that some could benefit. — Sohn Conference Panel.
- AI seat contraction: He argued that hiring freezes or staff reductions can reduce demand for per-seat software licenses; he did not establish a sector-wide margin collapse. — Invest Like the Best Interview.
- Concentration risk: He described focused technology portfolios and distinct long/short and long-only strategies; focus does not eliminate concentration risk. — Invest Like the Best Interview.
- Shorting valuation hype: His short criteria emphasize deteriorating competitive advantage and earnings expectations that he believes are too high. — Graham & Doddsville Interview.
Part 8: The Art of TMT Investing
- Foundational principles: His stated investment framework combines position on a technology S-curve, durable competitive advantage and underappreciated earnings power. — Graham & Doddsville Interview.
- The danger of linearity: He warns that straight-line forecasts can miss nonlinear earnings growth when a technology crosses an adoption threshold. — Invest Like the Best Interview.
- The value of specialization: He credits deep technology specialization with helping him understand adoption, competitors and industry structure. — Graham & Doddsville Interview.
- Hardware complexity: He argued at Sohn in 2026 that AI server racks and networking components were becoming more complex and innovation-intensive. — Sohn Conference Panel.
- Google's position: At Sohn in 2026, he expressed a bullish view that Google’s TPUs, model position and products gave it a strong AI opportunity; “won AI” was his investment opinion, not an established outcome. — Sohn Conference Panel.
- The 1998 parallel: He compared enterprises’ urgency to adopt AI with their need for websites during the late-1990s internet shift. — Invest Like the Best Interview.
- Holding through swings: He argues that deep knowledge of a company and a long-term thesis help an investor withstand short-term price swings. — Boston University Fireside Chat.
- The limits of financial data: His research tracks adoption and industry signals to anticipate future results rather than relying only on historical financial data. — Graham & Doddsville Interview.
- Ecosystem lock-in: He pointed to APIs, SDKs and orchestration products as parts of an AI-model ecosystem; their eventual durability remains uncertain. — Invest Like the Best Interview.
- Long-term focus: He seeks technology beneficiaries with durable advantages whose eventual earnings power the market has not fully recognized. — Graham & Doddsville Interview.