Decagon emerges from stealth to provide ‘human-like’ AI agents, transforming customer support for enterprises

“In the AI customer support chatbot space, we see more flashy demos than companies that can truly win the trust of an enterprise company,” Jesse Zhang, Decagon’s chief executive and co-founder, tells VentureBeat. It’s a bold statement for a startup just emerging from stealth and entering a market that has become increasingly crowded.

Founded by Zhang and Ashwin Sreenivas, Decagon provides companies with an AI-powered customer support platform that they claim mirrors human capabilities. Its solution seems to be resonating with the enterprise because the company says it has signed deals with Eventbrite, Bilt, Webflow, Rippling, Substack and other Fortune 1000 businesses in the span of less than a year.

How does Decagon work?

“Decagon empowers large enterprises and high-growth startups with the most human-like AI customer support agents that provide customers with contextualized, conversational interactions and CX leaders with greater control and visibility into their organizations,” Zhang says. According to him, having “human-like” AI agents means having a bot that services the entire customer support lifecycle beyond the conversational layer: “They have to reason through complex business logic, take actions, improve from feedback, tag conversations, analyze trends, write new articles, and more…We capture the entire customer support lifecycle in the most human-like ways possible.”

To help businesses, Decagon reveals it uses a combination of fine-tuned and third-party models to train its AI. It then ingests an organization’s knowledge bases and historical customer conversations while building custom integrations. This allows the startup’s agents to generate personalized customer responses and undertake actions on their behalf, such as processing refunds or postponing shipments. Ingesting past customer conversations will help the AI learn about the customer and develop custom responses instead of running through a decision tree and pulling a canned response. “[Customer experience] teams give it feedback just as if they have hired a new agent to the team. An agent that’s always on, responds immediately, and can scale infinitely,” Zhang explains.

As with other AI software providers, Decagon promises to give its users control and visibility into how its agents perform, analyzing conversations, reviewing tags and flagged anomalies, and making suggestions to their knowledge base to better address customer inquiries.

New funding support

Investors are noticing, too. In addition to its public debut, Decagon is disclosing for the first time it has $35 million in funding from a bevy of noted firms and individual venture capitalists. It received $5 million for its seed round led by Andreessen Horowitz and then raised a $30 million Series A round led by Accel, with participation from A*, Elad Gil, Box CEO Aaron Levie, Rippling Chief Operating Officer Matt MacInnis, Okta co-founder Frederic Kerrest, Lattice CEO Jack Altman, Klaviyo Ed Hallen, and others.

“Jesse, Ashwin, and the Decagon team have solved a critical pain point for enterprises: finding an AI product that goes beyond the customer-facing chat experience and addresses each layer of the customer operations stack from proactively updating an enterprise’s knowledge base to automatically routing insights through customer support to engineering and product teams,” Ivan Zhou, a partner at Accel, states.

The new capital will help further develop the company’s product while rapidly expanding its go-to-market and engineering teams.

A league of its own?

Although it boasts a list of notable businesses as customers, Decagon isn’t alone in using AI to tackle and improve how companies do customer support. It will undoubtedly continue to be an increasingly crowded marketplace as AI innovation advances. Businesses are moving away from chatbots in favor of more autonomous agents, even Digital Twins. Sprinklr, HubSpot and other providers have agents with similar capabilities to Decagon’s. Maven AGI, a generative AI platform for enterprise customer support that launched last month, raised $28 million in its debut and finds itself in a similar space.

However, Zhang believes his startup is “the leader in the new generation of generative AI customer support agents.” He admits there was pressure not to enter the space because of its competitive nature. However, “when we look at the makeup of this new generation of AI agents, many are unable to capture the complexity of an enterprise or take meaningful actions to solve customer pain points.” Zhang asserts that Decagon is the only company that “goes beyond the conversation to offer the most human-like experience across the entire lifecycle of a customer service interaction, from taking tangible action on behalf of the customer to providing CX leaders with valuable insights and trends regarding their knowledge base, and everything in-between.”