Coval accelerates AI agent development with automated testing for chat, voice, and other objective-oriented systems. Many engineering teams are racing to market with AI agents, but slow manual testing processes are holding them back. Teams currently play whack-a-mole just to discover that fixing one issue introduces another. At Coval, we use automated simulation and evaluation techniques inspired by the autonomous vehicle industry to boost test coverage, speed up development, and validate consistent performance.
External link for Coval
Ferry Building, 1
Suite 201
San Francisco, San Francisco, US
Coval reposted this
We're excited to join Deepgram for their AI Open House in San Francisco on October 22nd! 🎉 Coval and Deepgram work together to help enterprises evaluate and benchmark voice AI models. Join us alongside teams from Vapi, Daily, and LiveKit for an evening of great conversations about the future of voice AI. 📅 Wednesday, October 22, 2025 | 5:30 – 8:00 PM Signup link is in the comments!
Coval reposted this
🧵 This week in Conversational AI 🧵 Salesforce Dreamforce comes with great announcements in conversational AI! First of all, they introduced Agentforce Voice, a new feature in its AI-powered customer service platform that enables voice-based customer interactions, allowing AI agents to answer calls, customize tone and pronunciation, and integrate with major phone systems. Then – we hear that ElevenLabs joined the conference to talk about how they're part of powering the voice of Agentforce. And finally, Salesforce relaunched Slack as “the new agentic OS for work,” integrating AI agents, CRM data, and contextual intelligence into one workspace to transform how teams collaborate, make decisions, and boost productivity. What other Dreamforce announcements did I miss? Microsoft announced the introduction of sensitive data redaction in their Contact Center, enabling privacy-first voice AI experiences by automatically removing sensitive information from recordings, transcripts, and logs. ElevenLabs is hosting it'sElevenLabs Summit soon! On Nov 11, they're bringing together the leaders shaping the future of voice-first interfaces. Deepgram is hosting a massive happy hour in SF n October 22nd! We're excited to join the team for their AI Open House. Coval and Deepgram work together to help enterprises evaluate and benchmark voice AI models. Join us alongside teams from Vapi , Daily , and LiveKit for an evening of great conversations about the future of voice AI. Link in the comments! Reddit expanded its AI-powered conversational search feature, Reddit Answers, to five new languages: French, German, Spanish, Italian, and Portuguese. making it available in countries like Brazil, France, Germany, Spain, and Mexico. Walmart × OpenAI partner on a generative AI shopping assistant powered by Voicee LLC’s patented voice technology, ushering in the era of conversational commerce and validating foundational IP in voice-driven AI. Let me know what else happened this week!
Coval reposted this
The real speed hack for conversational AI? Evals. Andrew Ng just called it: the no.1 predictor of how fast you ship AI agents isn't your model choice or your tech stack - it's having disciplined evals and error analysis. Here's the problem: Voice and chat agents fail in way more ways than traditional ML. Latency issues. Bad interruption handling. Off-script responses. Workflow breaks. Without proper testing, you're flying blind, discovering bugs in production, and iterating in slow motion. This is where Coval changes the game: 1. Ship faster Stop building eval infrastructure from scratch. We give you audio metrics, LLM judges, workflow verification; everything you need to test comprehensively from day one. Teams go from prototype to production-ready in weeks instead of quarters. 2. Iterate with confidence Simulate hundreds of realistic conversations overnight. Spot regressions instantly. Know exactly what changed and why. No more "let's deploy and see what happens." 3. Turn testing into selling (!) Here's the bonus: your evaluation results become your best sales tool. Share public dashboards showing 95%+ resolution rates. Let prospects see exactly how your agent handles edge cases. Data-driven demos close deals. 4. Keep improving in production Push live calls through the same metrics. Catch issues early. Feed real edge cases back into your test suite. The feedback loop never stops. Bottom line: Teams using Coval ship 3-5x faster because they're not guessing but they're measuring and fixing what matters. Like Andrew says: Focus on systematic evaluation...That's how you win.
Coval reposted this
I met Al Gore!!! And guess what? He's super passionate about voice AI! Not a surprise to me, though. Voice AI has the opportunity to solve many urgent systems-level challenges and opportunities in the next 5 years. The big question is whether voice will just be a better interface or if it will actually become the dominant intelligent layer mediating all human-machine interaction. I bet you know where I stand on this :-) It's been an absolute honor to be part of the discussion on this topic with a group of leading voice AI providers, buyers, and decision-makers. Scaling Voice AI safely has been a major focus of mine, and having Coval be part of this conversation confirms that we're working on something that truly matters. As an independent provider, Coval is enabling voice AI companies to scale safely and reliably while acting as an independent trust layer. We've got your back when it comes to scaling! I'm thrilled to be part of this incredible journey, and I can't wait to see where voice AI takes us in the future!
Coval reposted this
So awesome to host 100+ people in our brand new SF office in FiDi for #sftechweek! Thank you for sharing your insights on “The State of Voice AI”: -Corinne Marie Riley, Partner @ Greylock -Karan Goel, CEO @ Cartesia -Kwindla Hultman Kramer, CEO @ Daily -Brooke Hopkins, CEO @ Coval -Daniel Hoske, CTO @ Cresta It was great to see the community come together and chat through what’s overhyped and underhyped in voice, how to get to production level AI agents, and more. We all need to continue to learn together in this quickly evolving space. Leaves me even more fired up for our first customer conference coming up in four weeks, Cresta Wave!
Coval reposted this
Awesome energy at the Coval Daily Langfuse Boundary (YC W23) AI Tinkerers Tavus Y Combinator Voice AI Hackathon today!!
Coval reposted this
There's a common narrative in the AI space that the foundational model providers will swallow up most startups, but I don’t think that applies to Voice AI. The idea overlooks how different voice really is at the application layer. Voice systems aren't just about model size: latency, accent diversity, privacy, integration with existing workflows, regulatory compliance, emotional nuance, and cost per interaction all matter in very domain-specific ways. Some startups will win by focusing on verticals like customer service, healthcare, or automotive, specializing conversational flows, localizing for languages and accents, and deeply integrating with enterprise systems. Big models may become cheaper and more capable, but they won't automatically replace the need for differentiated solutions built for particular user needs, sectors, or geographies. Foundation models are becoming commoditized infrastructure, which means the real value is shifting to everything around the model: deployment, UX, trust, reliability, and domain fit. It's like saying AWS killed all cloud startups because they offer servers. The infrastructure matters, but so does everything you build on top of it. #AI #VoiceAI #Evals
Coval reposted this
From watching incredible launches at OpenAI DevDay to hosting the Voice AI community at our office... yesterday was exactly why I love SF. Loren Phillips and I talked about applying lessons from self-driving to voice AI – specifically around human alignment and denoising. Turns out the frameworks for building safe autonomous vehicles translate pretty well to building reliable voice agents. The highlight was definitely having michael j goh from Krew (a Coval customer!) jump on stage to share how they prioritize agent reliability for their end customers. Real-world insights from someone shipping in production hit different. Thank you to everyone who came out to SoMa!
Coval reposted this
Can your agent handle large-scale production or is it just only good at demos? Last call! Registration closes at 2PM PT today ⏰ We're diving into this question tonight at Coval HQ – how do you actually know your AI agent is ready for production? Doors open at 6:30 PM. The Voice AI community is showing up. You should too. Grab your spot before 2PM: https://lnkd.in/gQ9C-Rqi
Coval reposted this
Monday's gonna be packed and I'm here for it 🔥 Honestly, the energy around Voice AI right now is just insane. We love hosting these events because there's SO much happening in the space, and getting everyone in the same room to learn from each other? That's where the magic happens. It's an honor to be part of #SFTechWeek and share this momentum with everyone coming into the city. Let's make it a good one! About the event: We're doing a deep dive on conversational AI evals – we'll talk about our approach to benchmarking in voice AI, how to build your architecture as you scale, and how to use evals to gain trust with your customers. See you Monday 👋 Sign-up link in the comments!