One hot category in the emerging AI space is customer support, which isn’t surprising, really, when you consider the technology’s potential to reduce contact center costs while increasing scale. Critics argue that productive AI-powered customer support technology could drive down wages, lead to layoffs and ultimately deliver a more error-prone end-user experience. Proponents, on the other hand, say genetic AI will augment — not replace — workers while allowing them to focus on more meaningful tasks.
Jesse Zhang is in the supporter camp. Of course, he is a little biased. Along with Ashwin Sreenivas, Zhang was a co-founder Decagona productive AI platform to automate various aspects of customer support channels.
Zhang is well aware of how fierce the competition is in the AI customer support market, which includes not only tech giants like Google and Amazon but also startups like Parloa, Retell AI and Cognigy (which recently raised $100 million) . With an estimate, the sector could be worth $2.89 billion by 2032from $308.4 million in 2022.
However, Zhang believes both Decagon’s engineering expertise and go-to-market approach give it an advantage. “When we first started, the prevailing advice we got was not to pursue the customer support space because it was too crowded,” Zhang told TechCrunch. “Ultimately, what worked for us was aggressively prioritizing what customers wanted and keeping a laser focus on what customers would get value from. This is the difference between a real business and an impressive AI demonstration.”
Both Zhang and Sreenivas have technical backgrounds, having worked at both startups and larger technology organizations. Zhang was a software engineer at Google before becoming a marketer at Citadel, the market-making firm, and founded Lowkey, a social gaming platform that was acquired by Pokémon GO maker Niantic in 2021. Sreenivas was chief development strategist at Palantir before co-founding Computer vision startup Helia, which he sold to unicorn Scale AI in 2020.
Decagon, which primarily sells to “high-growth” enterprises and startups, is developing as many customer support chatbots as possible. The bots, driven by first- and third-party AI models, can be fine-tuned, able to absorb a business’s knowledge bases and historical customer conversations to gain a greater understanding of contextual issues.
“As we started building, we realized that ‘human bots’ entailed a lot, as human agents are capable of complex reasoning, taking actions and analyzing conversations in retrospect,” Zhang said. “From talking to customers, it’s clear that while everyone wants greater operational efficiency, it can’t be at the expense of the customer experience – no one likes chatbots.”
So how? is not Are Decagon bots like traditional chatbots? Well, Zhang says they learn from past conversations and feedback. Perhaps most importantly, they can integrate with other applications to take action on behalf of the customer or agent, such as processing a refund, categorizing an incoming message, or helping write a support article.
On the back end, companies get analytics and control Decagon’s bots and their conversations.
“Human agents are able to analyze conversations to notice trends and find improvements,” Zhang said. “The AI-powered analytics dashboard automatically reviews and tags customer conversations to identify topics, flag anomalies and suggest additions to their knowledge base to better address customer queries.”
Now, genetic AI has a reputation for being less than perfect — and, in some cases, morally compromised. What Zhang would say to companies wary that Decagon’s bots will tell someone to eat glue or write article full of plagiarized contentor that Decagon will train their internal models on their data?
Basically, he says, don’t worry. “Providing customers with the necessary guardrails and monitoring for AI agents was important,” he said. “We optimize our models for our customers, but do so in a way that ensures it is impossible to inadvertently expose data to another customer. For example, a model that generates a response for customer A would never have any exposure to data from customer B.
Decagon’s technology — while subject to the same limitations as any other AI-powered app — is attracting brand-name clients like Eventbrite, Bilt and Substack, helping Decagon break even. Major investors have also jumped on board the venture, including Box CEO Aaron Levie, Airtable CEO Howie Liu and Lattice CEO Jack Altman.
To date, Decagon has raised $35 million in seed and Series A rounds led by Andreessen Horowitz, Accel (which led the Series A), A* and entrepreneur Elad Gil. Zhang says the cash is being used to develop products and expand San Francisco-based Decagon’s workforce.
“A key challenge is that customers equate AI agents with previous-generation chatbots, which don’t actually do their jobs,” Zhang said. “The customer support market is saturated with legacy chatbots, which have eroded lost consumer trust. The new solutions of this generation must cut through the noise of the establishment.”
