DevRev launches Voice AI for customer support calls

DevRev has launched Voice AI for its Customer Agent support product, extending the company's shared organisational memory into live customer calls.

The launch adds voice to the same framework DevRev already uses for chat and email support. The tool draws on customer records, support tickets, orders, documentation, code and business applications so an AI agent can respond during a live call and, where needed, pass the conversation to a human agent without losing context.

Voice AI has become a crowded market as companies look beyond chatbots and email automation. DevRev is positioning its product around issue resolution rather than speech alone, arguing that as large language models improve, realistic conversation is no longer the main challenge.

Industry data cited by DevRev reflects that pressure. A Gartner survey published earlier this year found that 91% of customer service and support leaders were under executive pressure to implement AI. Gartner also reported that only 20% of organisations had reduced agent headcount because of AI. Forrester has predicted that about one-third of brands rolling out AI in self-service will fail, largely because their underlying data is not ready.

Manoj Agarwal, Co-Founder and President at DevRev, said the main technical challenge is not natural-sounding speech but access to the right business context during a call.

"Voice has become another interface for AI," said Manoj Agarwal, Co-Founder and President at DevRev. "The interesting engineering problem isn't making an agent sound more natural. It's giving an agent enough context to make the right decision while the customer is still on the line. In an enterprise, the answer rarely lives in one place. It lives across tickets, documentation, customer records, orders, engineering systems and business applications. If an agent can't reason within that context, it must eventually hand the conversation to someone who can. Shared organisational memory changes that by giving an agent enough context to resolve a problem instead of simply responding to it."

Shared memory

According to DevRev, its Computer product continuously synchronises information across enterprise systems while keeping existing permissions in place. With the voice launch, DevRev is applying that same shared data layer and the same workflows to phone support rather than creating a separate platform for voice interactions.

That matters because many voice systems still act mainly as front-end routing tools. DevRev said its approach allows a voice agent to determine what happened, explain the issue, provide an update, start a corrective workflow and document actions within a single conversation.

The product also supports multiple languages and records and transcribes every call for later analysis. It connects to existing enterprise telephony through SIP-based integration, allowing customers to add voice support without replacing their telephony systems or rebuilding existing workflows.

Product lead

Alongside the launch, DevRev said Jennifer Heape will lead the Voice AI product effort. Heape previously co-founded Vixen Labs and worked on conversational AI product strategy.

Heape said the focus should remain on solving the customer's problem during the support interaction.

"I joined DevRev because the company is focused on what customers actually care about: resolving their issue during the conversation with customer support," said Jennifer Heape, Voice AI Product Lead at DevRev. "A call can sound polished and still be a poor experience if the agent cannot reach the information needed to solve the problem. Customers do not care which system holds the answer. They care whether they have to repeat themselves, wait for a transfer or call back again. Bringing voice into Computer means the conversation can carry the full business context with it, and that is what makes DevRev's Voice AI so exciting."

Architecture focus

DevRev linked the launch to its broader argument that AI performance in enterprise settings depends as much on system architecture and data access as on the underlying model. The company pointed to results from Enterprise-Bench, its open benchmark for testing enterprise AI systems under production-style conditions.

According to DevRev, Computer achieved 94.3% accuracy on that benchmark while using four times fewer tokens per correct answer than Claude Code on the same tasks with the same model. The company argued that the result shows the impact of retrieval, reasoning and action layers built around enterprise data rather than raw model choice alone.

Founded in 2020, DevRev sells AI software that builds a knowledge graph from a customer's existing business systems and uses that information for enterprise search, employee service desks and self-service customer support. The company is backed by Khosla Ventures and Mayfield and is led by Co-Founder and Chief Executive Officer Dheeraj Pandey and Co-Founder and President Manoj Agarwal.

Voice AI is available across the agentic suite in Computer.