Skip to content
YC Root AccessYC Root Access

What It Actually Takes to Deploy a Voice Agent to a Fortune 500

Brooke Hopkins is the founder and CEO of Coval (S24), a simulation and observability platform for voice agents that helps enterprises test, monitor, and evaluate AI-powered phone systems at scale — working with customers like Perplexity and Deepgram to process tens of millions of calls per month — and has just raised a $28.2M Series A. In this fireside, Brooke sat down with Harj Taggar, Managing Partner at YC to talk about how her years building evaluation infrastructure and developer tools at Waymo turned out to be surprisingly transferable to the world of voice agents, why voice is emerging as the first truly productionized use case for autonomous agents, and what it took to go from a broader evals idea to a deeply focused enterprise platform — including the moment a customer offered to pay her before she'd written a single line of code. https://www.coval.dev Apply to Y Combinator: https://www.ycombinator.com/apply Work at a startup: https://www.ycombinator.com/jobs

Harj TaggarhostBrooke Hopkinsguest
Jun 24, 202630mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Scaling enterprise voice agents with simulation, evaluation, and observability infrastructure

  1. Coval provides simulation and observability so companies can test voice agents at scale before exposing real customers and monitor behavior across tens of millions of production calls.
  2. Enterprises are adopting voice agents quickly because existing call-center infrastructure (IVRs, SOPs, call flows) makes the jump to autonomous voice less radical than other agent categories.
  3. Voice agents introduce new failure modes—audio oddities, hallucinations, and compliance/security risks—making rigorous QA and continuous evaluation essential rather than optional.
  4. The next performance unlock is better controllability for real-time/speech-to-speech systems and improved coordination between specialized components (STT, reasoning, TTS) that share context effectively.
  5. Coval’s company journey hinged on strong customer pull from voice teams, an intentional early focus on enterprise scale, and founder-market fit from Hopkins’ Waymo simulation/evaluation tooling background.

IDEAS WORTH REMEMBERING

5 ideas

Enterprise voice is the first truly scaled autonomous-agent deployment category.

Hopkins argues voice is already “productionized” autonomy, because companies can slot agents into existing phone workflows and immediately run them at high volume.

Testing voice agents with real calls doesn’t scale—simulation is the missing infrastructure.

Manually calling an agent repeatedly is slow, expensive, and statistically weak; simulation enables broader coverage of edge cases before risking customer-facing failures.

Voice QA must expand beyond classic contact-center metrics.

Enterprises need to evaluate not just outcomes, but also workflow/tool-call correctness and audio quality factors like latency, interruptions, and background noise handling.

Word error rate is often overvalued relative to intent and task success.

A conversation can tolerate imperfect transcription if the system correctly infers intent and completes the workflow, similar to humans understanding garbled audio on calls.

Agents fail in more “egregious” and brand-risky ways than humans.

Examples include vocal hallucinations (screaming/whispering/voice shifts) and confidently wrong statements—issues that require new guardrails and monitoring.

WORDS WORTH SAVING

5 quotes

Coval is a simulation and observability platform for voice agents, so we help you to scale your voice agents over millions of conversations so that you don't have to test your voice agents with real customers in production, and then also when you deploy your agents to production, you know what's happening out there in the wild.

— Brooke Hopkins

I think in two years or one year from now, it's going to be unacceptable to call an airline and be on hold for 20 minutes.

— Brooke Hopkins

Famously, voice agents will accidentally scream, or they'll start to whisper, or they'll change voices halfway through.

— Brooke Hopkins

It's been really cool to watch Coval grow because now I feel what product market fit feels like, where people are chasing us down to book meetings to push things through procurement. They're putting us on their back and just carrying us through the procurement process.

— Brooke Hopkins

So it's pretty easy to make something work once, but then to make it work over time is the challenging part.

— Brooke Hopkins

Simulation before production rolloutProduction observability for voice agentsEnterprise adoption drivers (IVR/SOP leverage)Voice-agent failure modes (audio, hallucinations, workflow errors)Evaluation strategy and trust in metricsWord error rate vs intent/task completionCascaded vs real-time speech-to-speech architectures

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

iOSAndroidClaudeChrome