Before Pat Grady co-led Sequoia Capital—before the $250 billion portfolio, before the ServiceNow and Zoom investments that defined enterprise software's cloud transition—he spent his days making 50 cold calls. The year was 2007, and Grady had just joined Sequoia at age 24 after three years at Summit Partners. His job description was straightforward: find companies. Call founders. Get meetings. Build relationships.
In August 2024, Grady posted a screenshot on X that revealed the unglamorous reality behind venture capital's carefully curated origin stories. The image showed Sequoia's CRM record for ServiceNow, starting in 2008. The entries documented persistence bordering on stubbornness: "left message with assistant," "sent email," "left voice mail," repeated across weeks. Multiple attempts to reach Fred Luddy, ServiceNow's founder, before finally securing a first response.
"My partners at Sequoia like to tell a story about how we had this brilliant SaaS thesis that led us to Fred Luddy, founder of ServiceNow," Grady wrote. "But the truth is that we pinged him because it was a company, and my job was to find companies. Here is the actual CRM record."
The admission contradicts venture capital's preferred narrative—that successful investments stem from thesis-driven research, pattern recognition, and strategic foresight. Instead, Grady's ServiceNow story reveals a more prosaic reality: discipline, repetition, and the law of large numbers. Make 50 calls daily. Have 200 conversations monthly. Over years, the volume generates outlier opportunities that retrospectively appear inevitable.
ServiceNow would eventually go public in 2012, reaching a $4.8 billion valuation. By November 2025, the company's market capitalization exceeded $220 billion, making it one of the most valuable enterprise software companies in the world. Sequoia's early investment—won through persistent cold calling rather than brilliant thesis work—generated billions in returns. The lesson shaped Grady's investment philosophy: execution beats strategy, persistence beats brilliance, and volume creates luck.
But the journey from Wyoming roofer to Sequoia co-steward required more than cold calls. It demanded navigating Sequoia's generational leadership transitions, surviving the firm's most challenging period in decades, and betting correctly on the biggest technological shift since cloud computing—artificial intelligence's transformation from research curiosity to trillion-dollar market opportunity.
Pat Grady grew up in Wyoming's Powder River Basin, a region known for coal mining, cattle ranching, and vast empty spaces. The area's economic identity centers on extraction industries—coal, natural gas, uranium—creating a culture far removed from Silicon Valley's startup ecosystem. His first job was roofing houses, physical labor that paid for college expenses and instilled work ethic lessons that would later inform his venture capital career.
The intellectual trajectory from Wyoming to Boston College occurred through academic achievement. Grady earned a Presidential Scholarship, a prestigious merit-based award covering full tuition. He graduated summa cum laude in 2004 with a Bachelor of Science in economics and finance, concentrating in mathematics. The quantitative training would prove essential for evaluating enterprise software business models—understanding SaaS metrics, cohort retention curves, net revenue retention rates, and unit economics that separate durable businesses from growth-at-all-costs fantasies.
Grady supplemented his Boston College education with a summer certificate in Advanced Econometrics and Game Theory from The London School of Economics and Political Science. The game theory training provided frameworks for analyzing competitive dynamics, strategic positioning, and network effects—concepts directly applicable to enterprise software markets where winner-take-most dynamics reward companies that achieve market leadership.
Before graduation, Grady interned with Citigroup's Healthcare team, gaining exposure to investment banking's financial modeling and valuation methodologies. But banking's focus on analyzing public companies didn't satisfy his interest in earlier-stage growth. In 2004, he joined Summit Partners as an Associate, entering the growth equity world that would define his career.
Summit Partners operates at venture capital's growth stage, investing $10 million to $500 million in companies with proven business models, consistent revenue growth, and clear paths to profitability or liquidity. The role positioned Grady between venture capital's early-stage risk-taking and private equity's mature company optimization. He learned to evaluate companies with real revenue, paying customers, and operational track records rather than PowerPoint projections and market size estimates.
The job requirements included those 50 daily cold calls and 200 monthly conversations. The volume-based approach to sourcing deals trained Grady in pattern recognition across hundreds of company pitches, thousands of financial models, and countless founder conversations. After three years, in 2007, Sequoia Capital hired him at age 24 to join its growth investment practice.
Grady joined Sequoia during the firm's transition from Don Valentine's founding era to Doug Leone's leadership period. Leone, who joined Sequoia in 1988, represented the second generation of Sequoia partners after Valentine's retirement. The firm's culture emphasized apprenticeship—younger partners learning from senior investors through shared deals, board observation, and decades-long mentorship relationships.
Roelof Botha, who joined Sequoia in 2003 from PayPal, was building the firm's growth-stage investment practice alongside Leone. Botha's operational background—PayPal's CFO during its hypergrowth phase and IPO—informed Sequoia's approach to later-stage investing. Unlike traditional venture capital, which often prioritizes growth over profitability, Botha brought discipline around unit economics, cash flow management, and capital efficiency from his PayPal experience.
Alfred Lin, who joined Sequoia in 2010 after serving as Zappos COO and chairman, brought complementary operational expertise. Lin's experience scaling Zappos from startup to Amazon acquisition (for $1.2 billion) provided frameworks for evaluating operational execution, company culture, and management team capabilities beyond pure technology or market opportunity assessment.
Grady's education came through observing and eventually co-investing alongside these partners. Sequoia's collaborative investment approach meant multiple partners diligenced deals together, attended board meetings jointly, and debated strategic decisions collectively. The structure transferred knowledge from senior to junior partners while reducing individual decision-making risk—one partner's miss could be caught by another's skepticism.
By 2015, eight years after joining Sequoia, Grady assumed leadership of the firm's growth-stage investing practice. The promotion reflected both his deal sourcing productivity and successful track record across multiple investments. His portfolio responsibilities would eventually include Amplitude, Drift, HubSpot, Okta, Qualtrics, ServiceNow, Snowflake, Zoom, Notion, and OpenAI—companies with combined market capitalizations exceeding $250 billion by November 2025.
The growth stage focus positioned Grady at enterprise software's most critical inflection point: when companies transition from product-market fit to market leadership. These investments require different skills than early-stage venture capital. Product and technology risks have been largely de-risked. The questions shift to execution, competitive positioning, and capital deployment efficiency. Can the company scale sales organizations? Build international operations? Maintain gross margins during growth? Defend market position against well-funded competitors?
Grady's quantitative background and Summit Partners training prepared him for these questions. By 2025, after 18 years at Sequoia, his investment approach combined disciplined financial analysis with relationship-building volume—the same cold-calling methodology that generated the ServiceNow opportunity but applied across thousands of companies, founders, and market opportunities.
Pat Grady's portfolio construction between 2008 and 2020 captured enterprise software's generational shift from on-premises installations to cloud-based Software-as-a-Service models. The transition fundamentally altered software economics, customer acquisition strategies, and competitive dynamics. Grady's investments—ServiceNow, Zoom, Snowflake, Okta, HubSpot, Qualtrics—represented the winning companies across different software categories during this architectural transformation.
ServiceNow pioneered the IT service management category in the cloud. Fred Luddy's insight was that enterprise IT departments needed ticketing, workflow automation, and service desk capabilities delivered through web browsers rather than installed software. The company went public in 2012 at a $2.5 billion valuation. By November 2025, ServiceNow's market capitalization exceeded $220 billion, making it one of the decade's most successful enterprise software investments. The company generated $9.3 billion in revenue for fiscal year 2024, demonstrating the massive markets addressable through cloud-based enterprise platforms.
Zoom represented the second wave of SaaS disruption—taking categories with incumbent on-premises leaders (Cisco WebEx, Microsoft Skype for Business) and rebuilding them as cloud-native applications with superior user experience. Eric Yuan founded Zoom after spending years at Cisco following its WebEx acquisition. His thesis: video conferencing could be radically simpler, more reliable, and more delightful than existing solutions. Sequoia invested in Zoom's growth rounds before the company's April 2019 IPO.
The COVID-19 pandemic validated Zoom's product superiority in spectacular fashion. The company's daily meeting participants exploded from 10 million in December 2019 to 300 million by April 2020 as global lockdowns forced businesses, schools, and families to conduct life remotely. Zoom's market capitalization peaked at $160 billion in October 2020. While the stock declined post-pandemic to approximately $60 billion by November 2025, the investment generated substantial returns for Sequoia and demonstrated Grady's ability to identify category-defining products before mainstream adoption.
Snowflake attacked the data warehousing category, competing against established players including Amazon Redshift, Google BigQuery, and legacy solutions like Oracle and Teradata. The company's architectural innovation—separating storage and compute in cloud data warehouses—enabled customers to scale analysis independently from data storage, reducing costs while improving performance. Snowflake went public in September 2020 in the largest software IPO in history, raising $3.4 billion at a $33 billion valuation. Warren Buffett's Berkshire Hathaway participated in the IPO, a rare endorsement of a unprofitable growth company by value investing's most famous practitioner.
By November 2025, Snowflake's market capitalization reached approximately $50 billion despite continued operating losses. The company generated $3.5 billion in annual product revenue with 110% net revenue retention, demonstrating that existing customers consistently expanded usage. The metric validated Snowflake's land-and-expand go-to-market strategy: sign customers with small initial deployments, deliver value through fast query performance and scalability, then capture budget expansion as customers migrate additional workloads to the platform.
Across these investments, Grady demonstrated pattern recognition around several key characteristics: large addressable markets with incumbent on-premises solutions ready for cloud disruption, founder-CEOs with deep domain expertise (Luddy from service management, Yuan from video conferencing, Snowflake's trio from data infrastructure), and product experiences measurably superior to existing alternatives. The portfolio construction wasn't luck—it was systematic evaluation of hundreds of companies against consistent investment criteria.
In 2021, Pat Grady led Sequoia Capital's investment in OpenAI, the artificial intelligence research laboratory founded in 2015 by Sam Altman, Ilya Sutskever, Greg Brockman, and others. The investment timing proved prescient. Just 16 months later, in November 2022, OpenAI released ChatGPT, triggering the generative AI revolution that would transform technology industry investment priorities and competitive dynamics.
The OpenAI investment represented a significant departure from Grady's previous portfolio. ServiceNow, Zoom, and Snowflake sold software to enterprise customers, generating recurring subscription revenue from day one. OpenAI operated as a research laboratory with unclear monetization pathways, burning hundreds of millions of dollars annually on GPU compute for training increasingly large language models. The business model risk was substantial—would enterprises pay for AI capabilities, or would open-source alternatives commoditize the technology?
Sequoia's investment memo, written by partner Sonya Huang with input from Grady and others, articulated the firm's thesis. The memo compared generative AI's potential to previous platform shifts—personal computers, the internet, mobile, cloud computing—and argued that AI would be bigger than all of them. The core insight: AI would not just create new software categories but fundamentally change how all software gets built and delivered.
The bet required conviction despite limited evidence. In 2021, OpenAI's GPT-3 model demonstrated impressive capabilities for text generation, but commercial applications remained experimental. The technical risks were significant—would scaling laws continue to improve model capabilities, or would diminishing returns limit AI progress? The competitive risks were equally daunting—Google, Microsoft, Meta, and other well-funded organizations were pursuing similar research directions.
ChatGPT's November 2022 launch vindicated the investment thesis spectacularly. The application reached 100 million users in two months, the fastest consumer product adoption in internet history. Microsoft accelerated its partnership with OpenAI, investing $10 billion and integrating GPT-4 across Azure, Office 365, and other products. OpenAI's revenue trajectory exploded from approximately $28 million in 2022 to $3.4 billion projected for 2024, demonstrating that enterprises would indeed pay premium prices for frontier AI capabilities.
By October 2025, OpenAI raised $6.6 billion in a funding round valuing the company at $157 billion post-money, making it one of the most valuable private companies in the world. Sequoia's 2021 investment generated paper returns exceeding 10x, potentially higher depending on the firm's ownership percentage and subsequent investment rounds. More strategically, the OpenAI relationship positioned Sequoia at the center of AI's development, providing access to frontier research, early product roadmaps, and insights into competitive dynamics across the AI ecosystem.
Grady's role in the OpenAI investment demonstrated evolution beyond SaaS pattern recognition. The bet required conviction in fundamental technology shifts rather than incremental business model improvements. It demanded comfort with business model uncertainty and capital intensity fundamentally different from typical software investments. The success validated Grady's ability to extend investment frameworks beyond proven categories into emerging technological paradigms—a critical skill for navigating AI's transformation of enterprise software.
In November 2025, Sequoia Capital announced that Pat Grady and Alfred Lin would become co-stewards of the firm, succeeding Roelof Botha in the leadership role Botha had held since Doug Leone's transition in 2022. The announcement marked Sequoia's return to a co-steward governance model after Botha's three-year tenure as solo steward.
The transition occurred during a complex period for Sequoia. The firm had navigated several significant challenges between 2022 and 2025: the collapse of FTX, in which Sequoia had invested $213.5 million (writing off approximately $200 million), the forced separation of Sequoia's China and India businesses into independent entities, and market volatility that severely impacted public market valuations of Sequoia's portfolio companies.
Botha's decision to step back from the steward role reflected Sequoia's traditional leadership succession model—transitions occur during periods of strength rather than crisis, ensuring continuity while refreshing strategic direction. Botha would remain a partner and board member of multiple portfolio companies, maintaining substantial influence while reducing day-to-day operational responsibilities.
The co-steward structure—Grady and Lin sharing leadership rather than a single managing partner—aligned with Sequoia's historical governance model. The firm operated under co-stewards during much of its history, including Don Valentine and Pierre Lamond, then Valentine and Mike Moritz. The dual leadership structure spreads decision-making authority, reduces single-point-of-failure risk, and ensures diverse perspectives inform strategic choices.
Grady and Lin brought complementary strengths to the partnership. Grady's 18-year tenure focused primarily on enterprise software and AI investments, building relationships with founders in developer tools, infrastructure, and application layer companies. Lin's background spanned e-commerce (Zappos), consumer (Airbnb), and increasingly AI investments (OpenAI, Replicate, Harvey). The combination provided coverage across technology's major investment categories.
The leadership announcement occurred alongside Sequoia's unveiling of two new funds totaling $950 million: a $750 million early-stage fund targeting Series A startups and a $200 million seed fund. The fund sizes nearly matched amounts raised three years earlier, signaling capital deployment consistency despite market volatility. The commitment to AI investments was explicit—Sequoia positioned artificial intelligence as "the biggest opportunity in venture capital's history."
Grady's elevation to co-steward formalized his role shaping Sequoia's AI strategy and portfolio construction across the technology stack. His public statements and conference presentations articulated a clear thesis: while foundation models like OpenAI attracted significant attention and capital, the application layer presented larger aggregate opportunities for venture returns. This positioning would define Sequoia's deployment strategy for the $950 million in new funds.
In October 2023, Roelof Botha articulated Sequoia's AI investment strategy in an interview that revealed the firm's positioning: "We're not actively seeking investments in companies involved in the development of AI technology," referring to foundation models. Instead, Sequoia was "focusing on AI applications that utilize foundation models."
By November 2025, Pat Grady had refined this thesis with specific capital allocation data. Speaking at Sequoia's AI Ascent 2025 conference, Grady disclosed that Sequoia had invested "an order of magnitude more dollars at the application layer, even though the revenue being generated at the application layer is a lot less." The statement revealed deliberate portfolio construction favoring applications despite foundation models' current revenue dominance.
The strategic reasoning reflected several insights. First, foundation model development requires massive capital—training runs for frontier models cost hundreds of millions of dollars, with compute requirements doubling approximately every six months. This capital intensity concentrates the market among well-funded players (OpenAI, Anthropic, Google, Meta) and reduces potential returns for investors—when companies raise billions at multi-billion valuations, even successful exits may generate modest multiples.
Second, foundation models face commoditization risk. As models become more capable, performance differences narrow. GPT-4, Claude 3.5 Sonnet, Gemini 1.5 Pro, and other frontier models demonstrate comparable performance across many tasks. Open-source models like Meta's Llama 3 and Mistral's offerings further compress pricing power. The commoditization dynamic suggests foundation models may become infrastructure—essential but low-margin—rather than capturing majority economic value.
Third, the application layer presents more diverse opportunities. Every software category faces AI-driven transformation: customer service (AI agents replacing ticketing systems), legal (contract analysis and research), healthcare (clinical documentation and coding), sales (automated prospecting and email generation), engineering (code generation and review), and hundreds of other verticals. The breadth creates room for multiple winning companies rather than winner-take-most dynamics common in infrastructure markets.
Sequoia's application layer portfolio reflected this thesis. The firm invested in Harvey (legal AI), Dust (enterprise AI assistant), Glean (workplace search), Replicate (AI model deployment), Hugging Face (AI developer platform), and Character.AI (conversational AI). Most recently, Sequoia led an investment in Rogo Technologies, a startup developing AI tools for investment banking analysts—directly attacking knowledge worker productivity in high-value industries.
Grady's public statements emphasized a specific pattern for successful AI applications: "Trust is the critical design pattern most AI companies miss." He observed that most AI products achieve 80% functionality quickly but the final 20% takes much longer and builds actual trust. The insight reflected lessons from portfolio companies—enterprises adopt AI cautiously, starting with low-risk use cases and expanding only after validating accuracy, reliability, and safety.
The trust framework explained why certain categories advanced faster than others. Customer service and sales automation, where mistakes carry limited downside, saw rapid AI adoption. Healthcare and legal applications, where errors could cause patient harm or legal liability, required extensive validation, regulatory approval, and professional oversight. Sequoia's portfolio balanced quick-adoption categories (Harvey, Glean, Dust) with longer-validation opportunities (healthcare AI remains underweight in Sequoia's disclosed portfolio).
Another Grady observation: "The 'data flywheel' appears in 100% of AI pitches but only 1% of companies actually demonstrate it works." The data flywheel hypothesis suggests that AI products improve through usage—user interactions generate training data, which enhances model performance, which attracts more users, creating a virtuous cycle. While theoretically compelling, Grady's skepticism reflected practical challenges: users don't tolerate poor initial experiences waiting for flywheels to spin, competitive moats require proprietary data rather than generic usage logs, and privacy regulations limit data collection and sharing.
By November 2025, Sequoia's AI portfolio spanned approximately 70 companies, with 17 remaining in stealth mode. The firm had invested roughly $150 million specifically in foundation model companies (OpenAI, Safe Superintelligence Inc., xAI) while deploying multiples of that amount across application layer opportunities. The capital allocation demonstrated conviction in the application layer thesis despite foundation models' current revenue and attention dominance.
In March 2025, Sequoia Capital hosted its second annual AI Ascent conference, convening founders, investors, and technology executives to discuss artificial intelligence's commercial trajectory. Pat Grady delivered the keynote presentation, outlining Sequoia's updated thesis on AI's market opportunity, adoption dynamics, and competitive evolution.
The headline claim positioned AI as venture capital's largest opportunity: "Nature hates a vacuum. There is a tremendous sucking sound in the market right now for AI... you are in a run like heck business right now." Grady's language deliberately invoked urgency—the window for capturing AI market share would not remain open indefinitely as competitive dynamics matured and market positions solidified.
Grady argued that AI was attacking both software and services markets simultaneously, representing "a profit pool at least an order of magnitude larger than previous technological transitions." The insight compared AI to prior platform shifts. Personal computers disrupted typewriters and calculators. The internet disrupted publishing and retail. Mobile disrupted cameras, GPS devices, and location-based services. Cloud computing disrupted on-premises IT infrastructure. Each transition addressed specific market segments.
AI, by contrast, could automate or augment knowledge work across every industry. The addressable market wasn't just software (roughly $1 trillion annually) but also services where human labor currently performs tasks AI could automate (tens of trillions annually). Grady described AI products evolving "from tools into copilots and ultimately autopilots, shifting from software budgets into labor budgets." The progression from assisting humans to replacing humans expanded the economic opportunity from technology spending to global GDP.
On adoption speed, Grady highlighted fundamental differences from previous technology transitions. Cloud computing required years of enterprise education about security, reliability, and total cost of ownership before achieving mainstream acceptance. Mobile apps faced distribution challenges—convincing users to download new applications, navigate app store discovery, and change behavior patterns. AI benefited from immediate awareness: "November 30th, 2022, ChatGPT comes out. The entire world is paying attention to AI."
The awareness advantage combined with existing distribution channels (web browsers, Slack integrations, API calls from existing software) and global connectivity (5.6 billion people online) to accelerate adoption. Grady's conclusion: "When the starting gun went off, there were no barriers to adoption." AI products could reach millions of users within weeks rather than years, compressing market formation timelines and intensifying competitive dynamics.
Grady also emphasized founder execution as the primary competitive moat: "The greatest moat in AI isn't data or tech—it's founders with relentless execution." The statement reflected skepticism about technological moats in the foundation model era. When multiple companies access similar base models (GPT-4, Claude 3.5, Gemini 1.5), differentiation shifts from model capabilities to execution variables: product design, distribution strategy, customer success, iteration speed, and talent density.
The AI Ascent conference itself served strategic purposes beyond education. By convening the AI ecosystem's power players, Sequoia reinforced its position as the AI industry's kingmaker—the firm where founders wanted capital, investors sought insights, and companies pursued partnerships. The conference created network effects around Sequoia's brand, generating deal flow, competitive intelligence, and strategic relationships that compounded across investment cycles.
In June 2023, Sequoia Capital announced it would divide its global business into three independent entities: Sequoia Capital (U.S. and Europe), HongShan (China), and Peak XV Partners (Southeast Asia and India). The split would complete by March 2024, ending Sequoia's unified global brand and operational structure that had existed since the firm's expansion into China in 2005 and India in 2006.
The announcement cited multiple factors. "It has become increasingly complex to run a decentralized global investment business," the official statement explained. The firm noted portfolio conflicts as local companies competed globally, creating situations where Sequoia faced pressure to choose between backing competitive companies in different geographies or passing on lucrative opportunities.
The unstated driver was geopolitical pressure. U.S.-China tensions had escalated dramatically between 2018 and 2023, encompassing trade wars, technology export controls, restrictions on Chinese companies accessing U.S. capital markets, and Congressional scrutiny of American venture capital funding Chinese AI and semiconductor companies. The Biden administration imposed restrictions on U.S. investments in Chinese AI, semiconductor, and quantum computing companies in August 2023, just months after Sequoia's split announcement.
For Pat Grady and Sequoia's U.S. partnership, the split eliminated substantial complexity. Sequoia China, led by Neil Shen, had become one of China's most successful venture firms, backing ByteDance (TikTok's parent company), Meituan, DJI, and other multibillion-dollar companies. But the success created conflicts—when Sequoia U.S. invested in American companies competing against Sequoia China's portfolio, how should the firms navigate competitive intelligence sharing, strategic guidance, and resource allocation?
The split also clarified capital allocation for limited partners—the pension funds, endowments, and sovereign wealth funds that invest in Sequoia's funds. Following the split, limited partners investing in Sequoia U.S. funds knew their capital would deploy exclusively in American and European opportunities, while those investing in HongShan funds would deploy in Chinese markets. The separation reduced geopolitical risk for American institutional investors facing potential regulatory or political backlash for indirect exposure to Chinese technology companies.
By November 2025, the split appeared strategically prudent. U.S.-China technology decoupling accelerated across semiconductors, AI, and cloud infrastructure. American companies faced increasing restrictions on Chinese market access, while Chinese companies confronted barriers to American technology and capital. Sequoia's early separation avoided forced divestments, regulatory penalties, or brand damage from potential future geopolitical conflicts.
For Grady personally, the split concentrated his responsibilities on U.S. and European markets—already the primary focus of his 18-year investment track record. His portfolio companies (Zoom, Snowflake, ServiceNow, OpenAI) operated primarily in Western markets, though many generated significant revenue from international customers including China. The geographic simplification enabled sharper focus on AI opportunity assessment and capital deployment across Sequoia's new $950 million in funding.
Beyond capital deployment, Pat Grady's role at Sequoia involves board service and strategic guidance for portfolio companies. As of November 2025, Grady served on the boards of Grow Therapy (mental health platform), Watershed (carbon accounting), Harvey (legal AI), Pilot.com (bookkeeping automation), Cribl (data pipeline management), and Attentive (SMS marketing). He also serves as investor and business partner for Hugging Face (AI developer platform) and Notion (productivity software).
The board portfolio reflected consistent themes across Grady's investment focus. Harvey represented pure-play AI application in a high-value vertical (legal services). Watershed addressed enterprise sustainability reporting requirements, a market expanding due to regulatory mandates and stakeholder pressure. Pilot.com automated financial back-office functions using AI and offshore labor, targeting small and medium businesses willing to outsource bookkeeping. Attentive provided SMS marketing infrastructure, capitalizing on mobile-first customer engagement.
Board service in venture capital serves multiple functions. Directors provide strategic guidance on market positioning, competitive dynamics, and capital allocation. They facilitate executive recruitment, customer introductions, and partnership development through network connections. They offer perspective during crises—whether product failures, competitive threats, or leadership transitions. And they monitor company performance, flag emerging risks, and recommend strategic course corrections.
Grady's board approach emphasized relentless execution—the same framework he advocated publicly at the AI Ascent conference. Multiple portfolio founders have described Grady's focus on execution metrics: customer acquisition costs, payback periods, net revenue retention, gross margins, and other quantitative indicators of business health. The discipline reflected Grady's Summit Partners training in growth equity, where companies have enough operational history to measure performance systematically.
The Harvey board seat carried particular strategic importance for Sequoia's AI thesis validation. Harvey raised a $100 million Series C in January 2024 at a reported $715 million valuation, positioning the company as one of the fastest-growing legal AI applications. The company achieved this valuation despite limited revenue, reflecting investor conviction in legal AI's massive market opportunity and Harvey's early leadership position.
But Harvey's success also illustrated AI application challenges. Legal research and contract analysis seem ideally suited for AI—they involve analyzing large text corpora, finding relevant precedents, and synthesizing information. Yet adoption required extensive validation because legal errors carry malpractice liability. Harvey addressed this through a "co-pilot" model where AI assists lawyers rather than replacing them, maintaining human accountability while improving productivity.
By late 2025, Harvey had expanded from legal research into contract analysis, due diligence automation, and regulatory compliance—use cases where legal teams could validate AI outputs before client delivery. The expansion demonstrated a pattern Grady had identified: AI applications succeed by starting with narrow, low-risk use cases where humans can verify outputs, then expanding into adjacent higher-risk applications as trust builds through demonstrated accuracy.
Sequoia Capital's leadership in AI investing faces competition from other elite venture firms deploying billions into the technology stack. Benchmark Capital, Andreessen Horowitz (a16z), Kleiner Perkins, Greylock Partners, and other established firms are racing to capture the most promising AI startups before valuations compress or market positions solidify.
Benchmark's concentrated portfolio strategy contrasts sharply with Sequoia's approach. Benchmark typically makes 6-8 new investments annually across its entire partnership, providing full attention and board representation to each portfolio company. Sequoia invests more broadly, with dozens of new investments annually across seed, early-stage, and growth categories. Benchmark's Eric Vishria led Exa Labs's $85 million round at $700 million valuation and delivered a 2025 talk analyzing zero-to-$100M ARR growth trajectories for AI companies.
Vishria's analysis concluded that AI companies grow approximately 10x faster than previous software generations, justifying higher revenue multiples despite limited profitability. The thesis supported premium pricing for AI investments but created risks—if growth rates normalized or competitive dynamics intensified, valuations could collapse rapidly. Benchmark's concentrated approach meant portfolio outcomes depended heavily on each investment's success, creating pressure to identify genuine category leaders rather than fast-growing also-rans.
Andreessen Horowitz built substantial AI presence through early OpenAI and Anthropic investments. The firm led Anthropic's $450 million Series C in May 2023 and participated in subsequent rounds, accumulating significant ownership in what became OpenAI's primary competitor. The diversified exposure across competing foundation models (OpenAI through indirect holdings, Anthropic directly) positioned a16z to benefit regardless of which model achieved market leadership.
a16z also deployed capital aggressively in AI infrastructure (Databricks, Anysphere/Cursor) and applications (Harvey, Character.AI, EvenUp, OpenEvidence). The firm's strategy emphasized full-stack exposure—foundation models, infrastructure, and applications—capturing value at multiple layers as AI markets matured. Managing Director Anjney Midha led many AI investments, partnering with Marc Andreessen and Ben Horowitz on firm-level AI strategy.
Kleiner Perkins, the venerated firm that backed Google, Amazon, and Genentech, reemerged as an AI player through investments in Hippocratic AI ($141 million Series B) and other healthcare AI companies. The firm's strategy focused on heavily regulated verticals—healthcare, financial services, government—where trust requirements and compliance burdens created barriers to entry for fast-moving startups but offered defensibility for companies that successfully navigated regulatory approval.
Greylock Partners pursued enterprise AI applications through Saam Motamedi's investments in intelligent application and AI infrastructure companies. Greylock's thesis centered on business model transformation—from seat-based to usage-based to outcome-based pricing—as AI shifted value capture from software licenses to measurable business outcomes.
For Pat Grady and Sequoia, competitive positioning depended on multiple factors beyond capital availability. Brand value—Sequoia's 50-year history backing Apple, Google, Airbnb, and other category-defining companies—attracted founders seeking validation and mentorship beyond money. Network effects from portfolio companies created cross-selling opportunities, shared insights, and ecosystem lock-in. And Sequoia's global platform (despite the China-India split) offered international expansion support that smaller firms couldn't match.
The AI deployment race would determine venture capital's next generation of leadership. Firms that captured the most promising AI companies at reasonable valuations would generate returns justifying future fundraising and talent retention. Those that overpaid or backed losing technologies would face limited partner skepticism and competitive disadvantage. For Grady, co-stewarding Sequoia during this transition meant ensuring the firm maintained its position as the AI industry's most connected and influential investor.
In early 2025, Sequoia Capital published its annual AI predictions, outlining the firm's expectations for technology development, market formation, and competitive dynamics. The document, informed by Grady's portfolio experience and broader partnership discussions, provided insight into Sequoia's investment thesis and deployment strategy.
The central theme: "If 2024 was the primordial soup year for AI, the building blocks are now firmly in place, and 2025 will be about sifting through those ideas to see which really work." The framing acknowledged that 2024's proliferation of AI startups, experiments, and product launches had created oversupply—thousands of companies pursuing hundreds of use cases with unclear validation of product-market fit, customer willingness to pay, or sustainable competitive positioning.
Sequoia's first prediction addressed foundation model competition. The firm expected intensified rivalry among language model companies as Claude, GPT, Gemini, and other models competed across performance benchmarks, pricing, and API reliability. Importantly, Sequoia predicted that performance gaps would narrow, reducing differentiation and compressing pricing power. The prediction validated Sequoia's application layer focus—if foundation models commoditize, applications capture the majority of economic value.
The second prediction identified AI search as a "killer app" gaining traction in 2025. The thesis reflected the emergence of Perplexity, Exa, and other AI-native search experiences that synthesized information rather than returning links. Traditional search engines (Google, Bing) faced disruption from interfaces that directly answered questions instead of requiring users to click through multiple websites. Sequoia's portfolio included companies building AI search infrastructure, positioning the firm to benefit from the category's growth.
The third prediction suggested AI capital expenditures would stabilize after 2024's explosive growth. Sequoia noted that foundation model training costs had escalated to hundreds of millions of dollars per training run, with leading labs spending billions annually on compute infrastructure. The firm expected this arms race to moderate as companies recognized diminishing returns from ever-larger models and shifted focus to more efficient architectures, better training data, and application-layer value capture.
The predictions document also highlighted several emerging patterns. AI adoption would shift from experimentation to production deployment as enterprises moved beyond pilots into scaled implementations affecting thousands of employees and millions of customers. Vertical AI applications (legal, healthcare, financial services) would demonstrate superior performance versus horizontal tools because vertical solutions could incorporate domain-specific knowledge, workflows, and compliance requirements. And open-source AI models would continue improving, creating price pressure on commercial model providers.
For Pat Grady, these predictions informed capital deployment priorities across Sequoia's $950 million in new funding. The application layer thesis suggested concentrating investments in vertical AI companies with defensible customer relationships, domain expertise, and early market leadership. The foundation model commoditization expectation reduced enthusiasm for late-stage model company investments at premium valuations. And the production deployment trend indicated that companies demonstrating real revenue traction and enterprise adoption would command premium valuations versus earlier-stage experiments.
In June 2024, Sequoia published an analysis titled "AI's $600B Question," examining whether AI revenue would justify the massive infrastructure investments flowing into GPUs, data centers, and foundation model development. The document, authored by partner David Cahn, quantified the gap between AI capital expenditures and revenue generation, raising concerns about sustainability and return on investment.
The core calculation: NVIDIA alone was projected to sell $150 billion worth of AI chips in 2024, implying total AI infrastructure spending (including data centers, power, networking, and cloud services) could reach $600 billion annually. To justify these capital expenditures through reasonable investment returns, the AI industry would need to generate approximately $600 billion in annual revenue—a figure far exceeding current AI product revenue across all categories.
The analysis provoked significant debate. Critics argued that Sequoia was undermining confidence in AI markets despite the firm's substantial AI portfolio, potentially damaging portfolio company valuations and fundraising prospects. Defenders noted that honest assessment of market risks served investors and entrepreneurs better than cheerleading unsustainable hype. The document reflected intellectual honesty about investment risks even while Sequoia continued deploying billions into AI companies.
For Pat Grady, the $600 billion question validated the application layer focus. Foundation model companies absorbed the majority of infrastructure costs—hundreds of millions for training runs, billions for compute capacity, ongoing expenses for inference serving. Application companies leveraged these models through API calls, avoiding direct infrastructure costs while capturing customer-facing revenue. If AI markets faced a revenue crisis relative to infrastructure spending, application companies with real customers and measurable ROI would survive while infrastructure and model companies faced margin compression.
By November 2025, evidence suggested the revenue gap was narrowing but remained substantial. OpenAI was projected to generate $3.4 billion in 2024 revenue, up from approximately $28 million in 2022—spectacular growth but still small relative to infrastructure investment. Microsoft reported AI services contributing to Azure's growth, though specific revenue remained undisclosed. GitHub Copilot surpassed $100 million in annual recurring revenue. Anthropic, Google's Gemini, and other model providers disclosed limited financial information.
The application layer showed more diverse revenue traction. Harvey, Glean, Dust, and other Sequoia portfolio companies reported enterprise customer growth and expanding annual contract values. But most remained unprofitable, investing revenue into customer acquisition, product development, and competitive positioning rather than optimizing for near-term profitability. The pattern echoed cloud computing's early years—when adoption trajectories mattered more than current profitability for venture-backed growth companies.
The $600 billion question also highlighted risks in Sequoia's portfolio construction. If AI markets failed to generate sufficient revenue, even application companies would face pressure as customer budgets tightened and enterprise buyers demanded clearer ROI justification. Grady's growth equity background emphasized sustainable unit economics, suggesting portfolio companies would need to demonstrate profitable customer cohorts before accessing growth capital for scaled expansion.
In July 2024, Pat Grady appeared on the "Invest Like the Best" podcast hosted by Patrick O'Shaughnessy. The episode, titled "Relentless Application of Force," explored Grady's investment philosophy, decision-making frameworks, and lessons learned across 17 years at Sequoia. The title encapsulated Grady's approach: consistent execution over prolonged periods generates compounding advantages that appear inevitable in retrospect but result from disciplined daily work.
Grady described Sequoia's investment process as collaborative rather than hierarchical. Unlike venture firms where individual partners control decision-making for their investments, Sequoia requires partnership consensus for new investments and major portfolio company decisions. The structure reduces individual autonomy but improves decision quality through diverse perspectives and collective pattern recognition.
The collaborative approach explained Sequoia's consistency across generational leadership transitions. When Doug Leone succeeded Don Valentine, when Roelof Botha succeeded Leone, and now with Grady and Lin succeeding Botha, the firm maintained cultural and strategic continuity because decisions reflect partnership consensus rather than individual leader preferences. The model prioritizes institutional knowledge transfer over individual genius.
On founder evaluation, Grady emphasized execution velocity over strategic elegance. "The greatest moat in AI isn't data or tech—it's founders with relentless execution," he stated at AI Ascent 2025. The framework reflected lessons from ServiceNow, Zoom, Snowflake, and other portfolio successes. Fred Luddy, Eric Yuan, and Snowflake's founding team shared intensity around customer obsession, product iteration, and organizational scaling that separated category leaders from well-funded failures.
Grady also discussed attribution and incentives within Sequoia's partnership. Unlike firms that assign economic credit to individual deal sponsors, Sequoia distributes carried interest across the partnership based on overall fund performance. The system reduces political infighting over deal ownership while encouraging partners to help each other's portfolio companies succeed. If one partner's company needs customer introductions, board expertise, or crisis management support, other partners contribute freely because collective success determines compensation.
The relentless application of force framework applied to Grady's own career progression. From cold calling 50 companies daily in 2007 to co-stewarding Sequoia in 2025, the trajectory resulted from consistent execution rather than singular breakthrough moments. The ServiceNow investment came from persistent outreach, not brilliant thesis work. The OpenAI investment came from building relationships across AI research communities over years, not opportunistic pattern recognition. The co-steward role came from 18 years of successful investments, board service, and partnership contributions, not political maneuvering.
Despite optimistic public positioning, Pat Grady faces substantial challenges stewarding Sequoia through AI's market maturation. The $950 million in new funds requires deployment into opportunities generating returns sufficient to justify Sequoia's premium position in venture capital. But multiple risk factors could undermine AI investment returns over the 7-10 year holding periods typical for venture-backed companies.
First, valuation inflation creates downside risk. AI companies routinely raise funding at revenue multiples exceeding 50x annual recurring revenue, levels historically associated with mature, profitable software companies rather than early-stage ventures. Harvey's reported $715 million Series C valuation on limited disclosed revenue exemplifies the dynamic. If revenue growth disappoints or competitive pressures intensify, down rounds and valuation resets could vaporize paper returns.
Second, AI commoditization threatens differentiation. As foundation models improve and open-source alternatives proliferate, application companies built on API-accessed models may lack defensible moats. If customers can swap underlying models without changing user experiences, pricing power shifts to model providers (who compete on price and performance) rather than application companies. Sequoia's portfolio would face margin compression and customer churn.
Third, enterprise adoption may disappoint revenue expectations. The $600 billion question remains unresolved—can AI markets generate revenue justifying infrastructure investment? If enterprises determine that AI productivity gains don't justify software spending, adoption slows and revenue projections collapse. Sequoia's application layer companies would face elongated sales cycles, reduced contract values, and increased churn as customers cut discretionary technology spending.
Fourth, regulatory intervention could restrict AI deployment. The European Union's AI Act, California's proposed AI regulations, and potential federal legislation could impose compliance burdens, liability frameworks, or capability restrictions that limit AI commercial viability. Healthcare AI faces FDA oversight, financial services AI confronts OCC and SEC scrutiny, and autonomous systems face transportation and safety regulations. Sequoia's portfolio companies would absorb compliance costs and face delayed market entry.
Fifth, geopolitical fragmentation might balkanize AI markets. U.S.-China technology decoupling already restricts American AI companies' Chinese market access and limits Chinese AI deployment in U.S. enterprises. If this pattern extends to Europe (through data sovereignty requirements) or other regions, total addressable markets shrink and global scaling becomes challenging. Sequoia's portfolio construction assumes winner-take-most dynamics in large, global markets—if markets fragment geographically or regulatory barriers create regional champions, venture returns suffer.
For Grady personally, these challenges test lessons learned during previous market cycles. The dot-com crash (2000-2002) demonstrated that revenue-free growth stories eventually face accountability. The 2008 financial crisis showed that even fundamentally strong businesses suffer during capital freezes. The 2022-2023 tech downturn proved that high-multiple software companies face severe valuation compression when growth slows. Each cycle reinforced focus on unit economics, capital efficiency, and sustainable competitive advantages over hype-driven narratives.
Grady's growth equity background emphasizes these fundamentals. Companies seeking Sequoia's growth capital need demonstrated business models, predictable revenue, and paths to profitability within reasonable timeframes. The discipline filtered speculative opportunities from genuinely scalable businesses. Applied to AI investing, the framework prioritizes companies with real enterprise customers, measurable ROI, and expanding use cases over experimental products with impressive demos but unclear monetization.
Pat Grady's elevation to co-steward positions him to shape Sequoia Capital's trajectory through the 2020s and potentially 2030s—a period likely to determine whether the firm maintains its position among venture capital's elite or yields market leadership to younger, more specialized competitors. The challenges extend beyond AI investment success to institutional evolution in a rapidly changing venture capital landscape.
Sequoia's historical advantages—brand prestige, network effects, operational support infrastructure—face erosion from multiple directions. Specialized AI firms like Radical Ventures, AIX Ventures, and Conviction offer deep technical expertise and focused portfolios that appeal to AI-native founders. Corporate venture arms (Google Ventures, Microsoft's M12, Salesforce Ventures) provide strategic partnerships and distribution channels that pure-play venture firms cannot match. Sovereign wealth funds and crossover investors deploy billions at later stages, competing for growth equity deals that historically represented Sequoia's differentiated strength.
Grady and Lin must navigate these competitive dynamics while preserving Sequoia's culture, maintaining partnership cohesion, and adapting strategy to technological and market evolution. The co-steward model distributes decision-making authority but requires alignment between leaders with different investment focuses, portfolio responsibilities, and strategic perspectives. Disagreements about capital allocation, partnership expansion, or fund strategy could create internal friction that undermines institutional effectiveness.
The China-India split removed complexity but also reduced Sequoia's global footprint. HongShan and Peak XV operate independently, eliminating information sharing, cross-border collaboration, and unified brand leverage that previously differentiated Sequoia from regionally focused competitors. American founders seeking Asian expansion now require separate relationships with independent firms rather than accessing Sequoia's integrated platform. The fragmentation may prove strategically costly if AI markets remain globally connected despite geopolitical tensions.
Generational transition within the partnership creates additional leadership challenges. Grady joined Sequoia in 2007; Lin joined in 2010. Both represent the generation after Botha (2003), who represented the generation after Leone (1988), who represented the generation after Valentine (1972). As Sequoia's founding and second generations exit active partnership roles, institutional knowledge transfer depends on deliberate mentorship, documented decision-making processes, and cultural preservation mechanisms. Without these, Sequoia risks becoming a brand without differentiated capabilities.
The success metrics for Grady's co-stewardship will emerge across investment cycles spanning 2025-2035. Did Sequoia's AI investments generate top-quartile returns relative to venture capital peers? Did the application layer thesis prove correct, or did foundation model and infrastructure investments deliver superior outcomes? Did Sequoia maintain deal access to the most promising AI startups, or did founders increasingly prefer specialized or corporate investors? Did the firm successfully navigate generational transition, or did key partners defect to launch competing firms?
These questions won't resolve for years—venture capital operates on decade-long timescales between investment and liquidity. But directional indicators will emerge sooner: portfolio company exit trajectories, fundraising success for subsequent Sequoia funds, partner retention and recruitment, and Sequoia's position in competitive deal processes. By 2027-2028, preliminary evidence will suggest whether Grady and Lin are building Sequoia's next 50 years or managing the gradual decline of an institution unable to adapt to AI-era venture capital.
Pat Grady's journey from Wyoming roofer to Sequoia Capital co-steward demonstrates how relentless execution compounds across decades into institutional leadership. The cold calls that generated the ServiceNow opportunity in 2008 established patterns—disciplined sourcing, relationship persistence, volume-driven luck creation—that produced Zoom, Snowflake, OpenAI, and portfolio companies with combined market capitalizations exceeding $250 billion by November 2025.
The co-stewardship appointment coincides with artificial intelligence's transformation from research curiosity to trillion-dollar market opportunity. Sequoia's thesis—that application layer investments will generate superior returns versus foundation models despite current revenue imbalances—positions the firm to capture value from AI's commercial maturation while avoiding infrastructure commoditization risks. The $950 million in new funding provides capital for deployment across this thesis, with Grady's 18-year track record informing portfolio construction and company selection.
But success is far from guaranteed. Valuation inflation, commoditization risks, adoption uncertainty, regulatory intervention, and geopolitical fragmentation all threaten AI investment returns. Competitive dynamics intensify as specialized AI firms, corporate venture arms, and crossover investors deploy billions into overlapping opportunities. And Sequoia itself faces institutional challenges—generational transition, partnership cohesion, and strategic adaptation to rapidly evolving markets.
The next decade will determine whether Pat Grady's "relentless application of force" philosophy—consistent execution over prolonged periods generating compounding advantages—proves sufficient for stewarding Sequoia through AI's transformation of technology markets and venture capital itself. The $250 billion portfolio provided credentials for leadership. The trillion-dollar opportunity creates both potential and risk. And the legacy question—whether Grady builds Sequoia's next 50 years or presides over its institutional decline—awaits resolution through investment outcomes that won't fully materialize until the 2030s.
For now, in November 2025, Grady occupies one of venture capital's most powerful positions: co-steward of its most prestigious firm, investor in its most transformative technology transition, and architect of strategies that will shape which AI companies achieve market leadership and generate the returns justifying venture capital's aggressive deployment into artificial intelligence's commercial future.