Decagon (AI company)
Last reviewed
Sources
No citations yet
Review status
Needs citations
Revision
v3 ยท 3,190 words
Improve this article
Add missing citations, update stale details, or suggest a clearer explanation.
Suggest edit
CtrlK
Last reviewed
Sources
No citations yet
Review status
Needs citations
Revision
v3 ยท 3,190 words
Add missing citations, update stale details, or suggest a clearer explanation.
Decagon is an American artificial intelligence company that builds AI agents for enterprise customer service, deploying conversational agents that resolve customer inquiries autonomously across chat, email, voice, and SMS. Founded in 2023 by Jesse Zhang and Ashwin Sreenivas, the San Francisco company raised a $250 million Series D in January 2026 that valued it at approximately $4.5 billion, and its agents had served more than 10 million end customers by that point.[1][2][7] Its platform integrates with business systems to take actions such as processing refunds, modifying orders, or escalating cases to human agents, and its named customers include Duolingo, Notion, Eventbrite, Substack, Oura Health, Chime, and Avis Budget Group.[3][4][7]
The company, headquartered in San Francisco with additional offices in New York City and London, emerged from stealth in June 2024 with a combined $35 million in seed and Series A funding from Andreessen Horowitz (andreessen horowitz) and Accel.[1][3] It subsequently raised three larger rounds within roughly nineteen months: a $65 million Series B led by Bain Capital Ventures in October 2024, a $131 million Series C co-led by Accel and andreessen horowitz in June 2025, and a $250 million Series D led by Coatue Management and Index Ventures in January 2026.[5][6][7] The Series D round valued Decagon at approximately $4.5 billion, bringing total disclosed funding to roughly $481 million; in March 2026 the company completed its first employee tender offer at the same $4.5 billion valuation.[7][15]
Decagon's reported customers include consumer and enterprise brands such as Duolingo, Notion, Eventbrite, Rippling, Substack, Bilt Rewards, Webflow, ClassPass, Oura Health, Chime, Avis Budget Group, Mercado Libre, Block, Deutsche Telekom, and Affirm.[4][5][6][7] It competes in a crowded category of AI-native customer experience platforms alongside sierra ai, Ada, Intercom's Fin, and Zendesk AI.[8] In a 2025 profile, Forbes characterized Decagon's strategic ambition as building an AI-native replacement for the incumbent customer service software stack, with Decagon and Sierra widely regarded as the two most prominent venture-backed challengers in that segment.[14]
Decagon is a conversational AI company whose product is an enterprise customer service platform powered by AI agent technology built on large language model systems. Rather than functioning as a scripted chatbot, Decagon's agents are designed to handle customer service interactions end-to-end: they answer questions from a company's knowledge base, reason through business-specific logic, and take actions in connected systems such as processing a refund or postponing an upcoming shipment.[3][4][12] Decagon describes its product as an "Agent Operating System" or AI agent platform for customer experience teams, and frames its long-term vision as giving every brand an AI "digital concierge" as its primary interface with customers.[2][4]
| Field | Detail |
|---|---|
| Founded | 2023[1] |
| Founders | Jesse Zhang, Ashwin Sreenivas[1][2] |
| Headquarters | San Francisco, California, U.S.[1] |
| Other offices | New York City; London[1] |
| Industry | Conversational AI, customer experience automation |
| Product | AI agents for enterprise customer service (chat, email, voice, SMS) |
| CEO | Jesse Zhang[2] |
| President | Ashwin Sreenivas[2] |
| Total funding (disclosed) | ~$481 million across Seed, Series A, B, C, and D[3][5][6][7] |
| Valuation (Jan 2026) | ~$4.5 billion[7] |
| Lead investors | Accel, Andreessen Horowitz, Bain Capital Ventures, Coatue, Index Ventures[3][5][6][7] |
| End customers served | 10 million+ (as of Jan 2026)[2][7] |
| Website | decagon.ai |
Decagon was founded in August 2023 by Jesse Zhang and Ashwin Sreenivas, who had previously been close friends and had each built and exited an earlier startup before joining forces on Decagon.[5][9] According to multiple accounts, the two reconnected and developed the idea for the company after meeting at a retreat hosted by Andreessen Horowitz (andreessen horowitz) in Utah.[16]
Jesse Zhang serves as Decagon's co-founder and chief executive officer.[2] He studied computer science at Harvard University (harvard university) and later founded Lowkey, a social platform for gamers to share gameplay clips, which participated in Y Combinator and raised venture funding from Andreessen Horowitz (andreessen horowitz).[10] Niantic, the augmented reality games company behind Pokemon Go, acquired Lowkey in late 2021 to bolster its social product features; Zhang joined Niantic's social team after the acquisition.[10] In November 2025 Forbes profiled Zhang in connection with Decagon's competitive positioning against Salesforce and sierra ai, describing him as a 28-year-old founder.[14] He was named to the 2026 Forbes 30 Under 30 list in the artificial intelligence category for his work at Decagon.[14]
Ashwin Sreenivas is Decagon's co-founder and, since 2025, has held the title of president; he was previously identified as chief technology officer in earlier company materials.[2][3] He holds bachelor's and master's degrees in computer science from Stanford University (stanford university) and earlier worked as a deployment strategist at Palantir Technologies.[11] In 2019 he co-founded Helia, a startup that applied deep learning to video understanding; scale ai acquired Helia in December 2020.[11] After the acquisition Sreenivas began working with Zhang on the ideas that became Decagon.[9]
Decagon was founded in San Francisco in August 2023, with an initial focus on building large language model-based AI agents that could handle customer service interactions end-to-end rather than functioning as scripted chatbots.[3][12] Index Ventures, which later co-led the company's Series D, has written that Zhang and Sreenivas did not begin with a fixed product idea, instead conducting extensive research and interviewing "hundreds of executives" before settling on customer experience as their initial market.[13] The company received a $5 million seed round led by Andreessen Horowitz (andreessen horowitz) shortly after founding, and built its first production deployments before publicly announcing itself in June 2024.[3] One profile of the company reported that Decagon reached seven-figure annual recurring revenue with only two people on staff by mid-2024.[12]
Decagon describes its product as an "Agent Operating System" or AI agent platform for customer experience teams.[2] The platform comprises AI agents that engage end customers across chat, email, voice, and SMS, along with tooling for the operations teams that build, monitor, and govern those agents.[2][3]
Decagon's agents are designed to autonomously handle conversational customer support across multiple channels. Beyond answering questions from a company's knowledge base, the agents integrate with backend systems and helpdesks to take actions on behalf of the customer, such as processing a refund or postponing an upcoming shipment.[3] Following its Series B round, the company stated that it was expanding into voice in addition to chat and messaging, and subsequent product materials confirm voice as a supported channel alongside chat and email.[2][5]
Decagon's platform includes a number of components documented on the company's site:
Third-party analyses have described Decagon's architecture as comprising a core conversational AI agent, routing logic to direct unresolvable inquiries to human agents, an agent-assist surface that brings relevant documentation to human reviewers within tools such as Zendesk, an administrative dashboard for cross-conversation insights, and a QA interface that exposes the agent's decision-making for review.[12]
According to coverage of the company's 2024 launch, Decagon's product combines third-party large language models with several of its own fine-tuned models, selecting between them based on the task at hand.[4] The company has described its agents as going beyond a "conversational layer" by reasoning through business-specific logic, taking actions in connected systems, and producing operational artifacts such as analytics summaries and draft knowledge base articles.[4] Decagon has emphasized transparency tooling that lets non-engineering staff inspect why an agent took a given action and adjust its behavior, a capability it positions as a differentiator against opaque, black-box agent platforms.[12]
Decagon uses outcome-based pricing rather than traditional per-seat SaaS pricing. It offers a per-conversation model, in which the customer pays a fixed price per end-customer interaction, and a per-resolution model, in which the customer pays only when an inquiry is resolved autonomously by Decagon's agent and escalations to human agents incur no charge.[12] Per third-party analysis of the company's go-to-market, the majority of Decagon's customers use the per-conversation pricing, which produces costs that scale predictably with overall support volume.[12]
In announcements tied to its 2025 Series C, Decagon reported that customers were achieving average deflection rates near 70 percent, with Duolingo reportedly above 80 percent deflection, Oura Health reporting roughly 3x increases in customer satisfaction, and ClassPass reporting a 95 percent decrease in support conversation costs.[6] Third-party analysis of the company has reported autonomous resolution rates of 70 to 90 percent across its customer base.[12] In published case studies, Curology is reported to have moved from approximately 5 percent to 80 percent autonomous ticket resolution after deploying Decagon, reducing operational costs by approximately 65 percent, while Rippling is reported to have reached an approximately 50 percent chat resolution rate after integrating Decagon with its internal APIs.[12] In the same Series C round, Decagon said it had grown from zero to "eight figures" in annual recurring revenue over the prior year and had more than quadrupled its customer base.[6] At the time of its January 2026 Series D, Decagon said it had added more than 100 new enterprise customers during the prior year and that its agents had served more than 10 million end customers.[2][7]
Decagon has publicly disclosed enterprise customers across several industries, with named references reported across the company's funding announcements and on its website:
Decagon's January 2026 Series D announcement said it served Fortune 100 enterprises across airlines, banking, telecom, and retail in addition to consumer-internet brands.[7] In a quote accompanying that round, Brian Choi, chief executive of Avis Budget Group, said: "With Decagon, we're moving customer engagement from reactive service to intelligent, concierge-led experiences that empower our frontline teams to resolve issues faster and more effectively."[7]
Decagon has raised five disclosed equity rounds, with cumulative announced funding of approximately $481 million.
Decagon raised a $5 million seed round led by Andreessen Horowitz (andreessen horowitz) shortly after its 2023 founding.[3] The round was not publicly announced until June 2024, when the company emerged from stealth alongside its Series A.[3]
On June 18, 2024, Decagon announced a $30 million Series A round led by Accel, bringing total disclosed funding to $35 million.[3] Participating investors included A*, Elad Gil, and a roster of operator angels, among them Aaron Levie (chief executive of Box), Howie Liu (chief executive of Airtable), Matt MacInnis (chief operating officer of Rippling), Aaref Hilaly (partner at Bain Capital Ventures), Mike Vernal (former Sequoia partner), Frederic Kerrest (co-founder of Okta), Jack Altman (co-founder of Lattice), and Ed Hallen (co-founder of Klaviyo).[3]
On October 15, 2024, Decagon announced a $65 million Series B round led by Bain Capital Ventures, with participation from Elad Gil, A*, Accel, BOND, and ACME Capital, bringing total disclosed funding to $100 million.[5] In its announcement, the company said the round had quadrupled its prior valuation and that it would use the capital to expand its engineering team, accelerate go-to-market, enter additional verticals, and add modalities including voice.[5]
On June 23, 2025, Decagon announced a $131 million Series C round co-led by Accel and Andreessen Horowitz (andreessen horowitz), valuing the company at approximately $1.5 billion.[6] Existing investors A*, Bain Capital Ventures, and BOND participated, alongside new investors Avra, Forerunner, and Ribbit Capital.[6] The company said the round drew approximately five times more investor demand than capacity.[6] Cumulative disclosed funding reached approximately $231 million.[6]
On January 28, 2026, Decagon announced a $250 million Series D round led by Coatue Management and Index Ventures, tripling its valuation to approximately $4.5 billion in under six months.[7] New investors included Coatue, Index Ventures, ChemistryVC, Definition Capital, and Starwood Capital, with existing investors Andreessen Horowitz, A*, Accel, Avra, Bain Capital Ventures, Elad Gil, Forerunner, Ribbit Capital, and T.Capital also participating.[7] Cumulative disclosed funding reached approximately $481 million.[7]
In the announcement, co-founder and chief executive Jesse Zhang said: "This fast growth and large capital investment are both testaments to why our fundamentally different concierge approach to product is winning the market over and disrupting the outdated incumbent strategies."[7] Lucas Swisher, a general partner at Coatue Management, said the company's platform "gives enterprises greater control, transparency, and speed in deploying AI agents at scale."[7]
On March 4, 2026, Decagon announced that it had completed its first employee tender offer, allowing its more than 300 employees to sell a portion of their vested shares at the company's $4.5 billion valuation from the Series D.[15] The secondary transaction was led by the same investors that had backed the Series D less than two months earlier, including Coatue, Index Ventures, Andreessen Horowitz, Definition Capital, Forerunner, and Ribbit Capital.[15] Zhang described the tender as an opportunity to combine recent investment demand and growth milestones with rewarding the team.[15] At the time of the tender, Decagon reported more than 100 large customers, including Avis Budget Group, 1-800-Flowers, Quince, Oura Health, and Away Travel.[15]
| Round | Date | Amount | Lead investor(s) | Valuation |
|---|---|---|---|---|
| Seed | 2023 | $5M | Andreessen Horowitz[3] | Not disclosed |
| Series A | June 2024 | $30M | Accel[3] | Not disclosed |
| Series B | October 2024 | $65M | Bain Capital Ventures[5] | Not disclosed |
| Series C | June 2025 | $131M | Accel; Andreessen Horowitz[6] | ~$1.5B[6] |
| Series D | January 2026 | $250M | Coatue; Index Ventures[7] | ~$4.5B[7] |
| Tender offer | March 2026 | n/a (secondary) | Coatue; Index Ventures[15] | ~$4.5B[15] |
Decagon competes in the rapidly growing market for AI-native enterprise customer service automation, which expanded significantly in 2024 and 2025 as large language model-based agents began to displace earlier rule-based chatbots.
The most frequently cited direct competitor is sierra ai, the conversational AI company co-founded in 2023 by bret taylor (former co-chief executive of Salesforce and chair of OpenAI's board) and Clay Bavor (a former Google executive), which pursues a similar enterprise AI agent strategy and reached a similar multibillion-dollar valuation in the same period.[8] Other commonly compared platforms include Ada, an AI-native customer service automation platform; Intercom's Fin, an AI agent layered on top of Intercom's helpdesk and operable on third-party helpdesks; and Zendesk AI, the native AI suite within Zendesk's customer service platform.[8] Older AI-driven customer service vendors such as ASAPP and Cresta also compete in adjacent enterprise segments.[8]
Comparative analyses generally position Decagon as a standalone AI agent platform aimed at large enterprises that are willing to replace or sit alongside their existing helpdesk, in contrast with vendors such as Fin or Zendesk AI that are tightly coupled to a specific helpdesk product.[8] A November 2025 Forbes profile framed the broader competitive contest as Decagon and Sierra both attempting to build the dominant AI-first replacement for incumbent customer service software providers including Salesforce Service Cloud, with each company having reached multibillion-dollar valuations within roughly two years of founding.[14]
In its 2026 class, Forbes named Jesse Zhang to its 30 Under 30 list in the artificial intelligence category, citing Decagon's rapid revenue growth and adoption by more than 100 enterprise customers.[14] In the same profile, Forbes reported that Decagon had reached approximately $10 million in annualized revenue in 2024 and had crossed approximately $30 million in annualized revenue during 2025.[14] Independent research firm Sacra similarly estimated that Decagon reached roughly $35 million in annualized revenue by October 2025, up from about $10 million at the end of 2024.[17] In its perspective accompanying the Series D investment, Index Ventures described the founders as "disciplined operators" and former math and science Olympiad competitors who had previously delivered successful exits as repeat founders, citing the team's "rigor, humility, and commitment to never stop learning" as a decisive factor in the firm's decision to co-lead the round.[13] Sofia Dolfe, a partner at Index Ventures, said the firm was investing in the founders "because they're reimagining customer experience from first principles."[7]
Imagine a really smart helper that answers a company's customer messages day and night. When you message a store or an app for help, sometimes a computer program answers instead of a person. Decagon makes those computer helpers, called AI agents, and they are good enough to actually fix your problem, like giving you a refund or changing your order, not just send you in circles. Big companies pay Decagon so their customers get fast answers, and investors have given Decagon a lot of money because its helpers work well.