One hot-button category within the generative AI space is customer support. That's not surprising, considering the technology has the potential to scale back call center costs while increasing scalability. Critics argue that generative AI-powered customer support technology could depress wages, result in layoffs, and ultimately create a more error-prone end-user experience. Proponents, however, say generative AI will complement—not replace—employees while allowing them to give attention to more meaningful tasks.
Jesse Zhang belongs to the camp of supporters. Of course, he’s a bit biased. Together with Ashwin Sreenivas, Zhang founded decagona generative AI platform for automating various facets of customer support channels.
Zhang is well aware of how competitive the AI-powered customer support market is, which incorporates not only tech giants like Google and Amazon, but additionally startups like Parloa, Retell AI and Cognigy (which recently raised $100 million). According to at least one estimate, The sector could reach a price of $2.89 billion by 2032in comparison with $308.4 million in 2022.
However, Zhang believes that each Decagon's technical expertise and go-to-market approach give the corporate a bonus. “When we began, the primary thing we were advised to not pursue was customer support since it was too crowded,” Zhang told TechCrunch. “Ultimately, what worked for us was aggressively prioritizing what customers want and specializing in what provides value to customers. That's the difference between an actual company and a flashy AI demo.”
Both Zhang and Sreenivas have technical backgrounds and have worked at each startups and bigger tech firms. Zhang was a software engineer at Google before becoming a trader at market-making firm Citadel and founding Lowkey, a social gaming platform that was acquired by Pokémon GO maker Niantic in 2021. Sreenivas was a deployment strategist at Palantir before co-founding computer vision startup Helia, which he sold to unicorn Scale AI in 2020.
Decagon, which sells primarily to enterprises and “high-growth” startups, develops chatbots for customer support. The bots are driven by first- and third-party AI models and are fine-tunable. They can ingest an organization’s knowledge base and historical customer conversations to achieve a greater contextual understanding of issues.
“When we began constructing, we realized that 'human-like bots' have so much to achieve, as human agents are able to complex reasoning, executing actions and analyzing conversations after the very fact,” Zhang said. “It's clear from conversations with customers that while everyone wants greater operational efficiency, it will probably't come on the expense of the client experience – no person likes chatbots.”
How are Decagon's bots just like traditional chatbots? Well, Zhang says they learn from past conversations and feedback. Perhaps more importantly, they integrate with other apps to perform actions on behalf of the client or agent, equivalent to processing a refund, categorizing an incoming message, or helping write a support article.
In the backend, firms get analytics and control over Decagon’s bots and their conversations.
“Human agents can analyze conversations to discover trends and find improvements,” Zhang said. “Our AI-powered analytics dashboard routinely reviews and tags customer conversations to discover topics, flag anomalies, and suggest additions to their knowledge base to higher answer customer queries.”
Generative AI has a fame for being imperfect – and in some cases ethically questionable. What would Zhang say to firms that fear Decagon’s bots will tell someone to eat glue or Articles filled with plagiarized contentor that Decagon will train its internal models using their data?
Essentially, he says, there's nothing to fret about. “It was essential to supply customers with the vital protections and monitoring capabilities for his or her AI agents,” he said. “We optimize our models for our customers, but we accomplish that in a way that ensures no data might be by chance exposed to a different customer. For example, a model that generates a response for customer A would never be exposed to data from customer B.”
Decagon's technology – which has the identical limitations as every other generative AI-powered app – has recently attracted big-name customers like Eventbrite, Bilt and Substack, helping Decagon break even. Big-name investors have also jumped into the corporate, including Box CEO Aaron Levie, Airtable CEO Howie Liu and Lattice CEO Jack Altman.
So far, Decagon has raised $35 million in seed and Series A rounds involving Andreessen Horowitz, Accel (which led the Series A), A* and entrepreneur Elad Gil. Zhang says the cash will go toward product development and expanding Decagon's San Francisco-based workforce.
“One of the most important challenges is that customers equate AI agents with previous generation chatbots that don't really do the job,” Zhang said. “The customer support market is saturated with older chatbots which have lost customer trust. New solutions of this generation must break through the noise of incumbents.”
