Before starting her company, Brooke Hopkins worked on systems that helped self-driving cars operate safely on public roads. At Waymo, her team ran millions of simulated miles for every code change, since failures after launch were not an option for autonomous vehicles. She saw that similar standards would soon be needed for AI voice agents, even though few people noticed this trend back then.
Today, Hopkins’ startup Coval raised $28 million in a Series A round led by Norwest, with Base10 Partners, Twilio Ventures, and Y Combinator also investing. This brings Coval’s total funding to $31 million since it launched in 2024.
“Every company is going to have a voice agent just like they have a mobile app or a web app. But today, most enterprises don’t have the infrastructure to deploy these systems with confidence,” Hopkins says.
More than $7 billion was invested in voice AI in the first quarter of 2026, and the market could grow to over $20 billion by 2031. Companies are quickly adding voice agents to customer service, healthcare, and financial services.
Still, most rely on manual testing, in which engineers review call transcripts, look for problems, and hope their fixes work. This method does not scale well.
Applying Waymo’s testing methods to phone calls
At Waymo, Hopkins led the team that built evaluation systems. They developed tools to run simulations on distributed computers, which helped Waymo launch cars on public roads. After she left, Hopkins talked to AI agent teams and found they faced similar problems: too many possible inputs, no reliable way to test, and serious consequences when things went wrong.
Coval, founded in 2024 through Y Combinator and based in San Francisco, runs tens of millions of simulated tests for voice agents. The platform checks for issues such as accents, interruptions, background noise, and unusual situations before real users are affected.
After launch, Coval continues to monitor agents and automatically sends failed calls back into testing. For example, a financial services company can simulate thousands of callers who give conflicting information or hang up unexpectedly before the agent talks to real customers.
The startup says its platform can reduce manual quality checks by up to 30x and speed up deployment by up to 10x.
Coval’s main competitors are Hamming, which handles regulatory edge cases for healthcare and financial services, and Roark, a YC W25 graduate that has processed over 10 million minutes of calls and specialises in replaying failed conversations with updated agent logic.
While these companies focus on certain areas, Coval provides a complete solution. It offers pre-deployment simulation, live monitoring, and human review, all built specifically for voice audio rather than adapted from general language model tools.
Zoom and Deepgram already use Coval
Coval works with Deepgram, Zoom, and more than 60 other companies. Endorsements from Deepgram and Zoom stand out because both have extensive experience with voice AI failures.
“Reliability and observability are a top priority for us at Zoom as voice AI moves into customer-facing production environments. Coval gives Zoom’s customers the ability to evaluate conversations systematically at scale, identify edge cases before they impact users, and move significantly faster with confidence,” says Ram Rajagopalan, head of product for CX AI, Zoom.
Deepgram, which provides voice AI infrastructure to other companies, uses Coval to test its products before they launch.
“Voice agents introduce a new level of complexity compared to traditional software testing. Brooke has built Coval into a core part of the modern enterprise evaluation stack by improving reliability prior to scaled deployment. For any serious enterprise deployment, this is no longer a nice-to-have,” adds Anoop Dawar, COO, Deepgram.
The investors and what’s ahead
Norwest, which has invested in companies like Gong, Vuori, and Spiff, which is now part of Salesforce, led the funding round. Scott Beechuk, a partner at Norwest, said they invested because of Hopkins’ experience at Waymo.
“With her deep experience building evaluation systems for autonomous technologies at Waymo, Brooke is uniquely positioned to lead Coval in defining how companies deploy and scale voice agents reliably. She helped prove self-driving cars could work, and now she’s tackling voice AI,” notes Beechuk.
Twilio Ventures is involved in more than just funding. Its enterprise customers, who use its voice infrastructure, are a natural fit for Coval’s evaluation platform. Twilio wants to make sure these deployments are successful.
“Trust is critical to scaling these experiences. Our investment in Coval reflects our conviction that comprehensive evaluation and testing tools, combined with a strong observability and reliability layer, are foundational to maintaining momentum in today’s voice AI renaissance,” says Andy O’Dower, VP and field CTO, Twilio
Coval plans to use the new funding to grow its sales and solutions engineering teams and improve its product by adding deeper simulations, new integrations, and better human review and monitoring. The company says its revenue has grown ten times year over year, but it has not shared its ARR or current headcount goals.
A big question for the industry is whether voice AI evaluation will stay independent or become part of the platforms it supports. Twilio’s choice to invest in Coval, rather than build its own tool, shows that at least one platform prefers to keep evaluation separate.
