Sequoia Capital has emerged as the top investor in agentic AI, the fast-growing category of software that can plan and execute meaningful work autonomously, while a separate AI startup backed by Sequoia's China affiliate just hit a $250 million valuation. The venture capital firm's portfolio includes some of the most valuable companies in the space, and its early conviction in these businesses before they became obvious winners sets it apart from competitors like Andreessen Horowitz, Benchmark, and Accel.

Why Is Sequoia Winning in Agentic AI?

Sequoia's strength in agentic AI comes from a combination of early entry, portfolio breadth, and repeated investment conviction. The firm led a $21 million Series A for Harvey, a legal automation platform, back in 2023, when enterprise AI agents still had limited commercial proof. Harvey later raised $200 million at an $11 billion valuation, with customers running more than 25,000 custom agents on the platform.

Similarly, Sequoia and Benchmark invested a combined $110 million into Sierra, an enterprise customer-service agent platform, around its launch in 2024. By May 2026, Sierra was raising $950 million at a post-money valuation above $15 billion, with more than 40 percent of the Fortune 50 as customers and agents handling billions of interactions.

"The real insight is that in this AI era, we don't have a content shortage. The opposite is true. We have too much content. We have too much video footage sitting on our computers that isn't being used," said Henry Kang, founder of Clipto.

Henry Kang, Founder at Clipto

Sequoia's portfolio extends across multiple domains within agentic AI, including Harvey for legal work, Sierra for customer interactions, XBOW for cybersecurity, Factory for coding, and Sail Research for long-running agent infrastructure. This breadth, combined with early entry before these companies became growth-stage winners, gives Sequoia a credible edge in a market still finding its footing.

What Makes Agentic AI Investment So Difficult to Rank?

Ranking agentic AI investors is unusually messy because the category has expanded faster than investment track records can mature. Over a 12-month period, the market saw 40 disclosed pure-play agentic AI rounds worth approximately $3.37 billion, with another 12 qualifying rounds worth roughly $676 million. However, the term "agentic AI" now describes several distinct markets that barely resemble one another.

The category now includes customer-service agents, legal automation, coding assistance, cybersecurity, workflow orchestration, browser infrastructure, memory systems, identity management, and payment processing. This fragmentation means that a large portfolio can hide very different levels of investor conviction and expertise. As a result, early entry, company quality, follow-on conviction, and commercial progress tell a more accurate story than deal count alone.

The technology is also moving from experiments into production. A 2026 survey by LangChain of more than 1,300 people working with AI agents found that 57 percent already had agents in production, and among those teams, 89 percent had implemented observability tools. This shift from proof-of-concept to real-world deployment means investors can now judge their choices based on actual commercial outcomes rather than theoretical potential.

How to Evaluate Agentic AI Investments Like a Venture Capitalist

• Early Entry Timing: The strongest investor records typically start before a company becomes an obvious growth-stage winner, such as Sequoia's 2023 entry into Harvey or Benchmark's seed-stage backing of LangChain, rather than later nine-figure checks after success is proven.
• Portfolio Breadth and Specialization: Investors with exposure across multiple agentic AI domains, from applications to infrastructure to security, demonstrate deeper conviction than those concentrated in a single vertical or relying on large general AI portfolios.
• Repeat Conviction and Follow-On Investment: Firms that continue investing in companies through multiple funding rounds signal genuine belief in the technology and management team, not just opportunistic capital deployment.
• Commercial Proof Points: Real-world metrics like customer adoption, revenue growth, and production deployment matter more than theoretical benchmarks, especially in a young category where venture returns have not yet materialized.

Benchmark stands out for having fewer relevant agentic AI bets than most other top-tier firms, yet that concentration is almost the point. Its positions in LangChain, 11x, and Gumloop give it one of the highest concentrations of strong early agent investments in the market. Accel, meanwhile, has expanded fastest, growing from a particularly strong Decagon position into orchestration, browser infrastructure, agent payments, workplace agents, and security through companies including n8n, Kernel, Sapiom, Viktor, and Tolmo.

What About Smaller AI Startups Outside the Agentic AI Spotlight?

While Sequoia and other top-tier firms focus on agentic AI, smaller startups are raising significant capital in adjacent AI markets. Clipto, a San Francisco-based startup with teams in Singapore and Hong Kong, just raised $15 million in an all-equity round at a $250 million post-money valuation. The company's investors include HSG, formerly known as Sequoia China, along with GL Ventures, EnvisionX Capital, Palm Drive Capital, Hans Tung, Lu Zhang, and 522 Ventures.

Clipto indexes videos, audio, images, meetings, and other files on a user's computer, allowing users to search by describing what they're looking for or asking AI tools like ChatGPT or Claude to find specific content. The company was founded in 2023 by Henry Kang, who spent roughly two decades working on related problems, including a PhD at Carnegie Mellon University in 2006 focused on robots that recorded their surroundings and identified objects.

Clipto's user base has expanded far beyond its initial target of video creators. While creators made up only about one-quarter to one-third of users at the time of the funding announcement, the rest include lawyers, doctors, researchers, marketers, human resources professionals, professors, and students. The company says more than 30 million people have used its products since launch and that it has hundreds of thousands of paying subscribers.

The startup reached $15 million in annual recurring revenue at the beginning of 2026 and remains profitable on a net-income basis, according to Kang. Clipto has just over 20 employees split among the San Francisco Bay Area, Hong Kong, and Singapore. About two weeks before the funding announcement, Clipto added support for Model Context Protocol, or MCP, a standard that allows AI applications to connect with outside sources of information.

Clipto is entering a market where larger companies like Adobe, Apple, and Google already control much of the data and software people use to manage it. Adobe offers AI-powered search in Premiere, while Apple Photos and Google Photos allow users to find photos and videos using natural-language descriptions. However, Kang noted that Clipto can search across videos, audio, images, and documents, while products from Adobe, Apple, and Google generally focus on files stored within their own services.

The new funding will go toward the AI models and computing infrastructure needed to run Clipto on consumer hardware, as well as integrations with more AI agents. All processing runs locally on the user's device without requiring cloud services, and access to indexed files requires the user's active request and authorization.