January 23, 2025
Congratulations to the @covaldev team on this milestone. I've learned a lot from @bnicholehopkins and her colleagues over the past few months.
I love that Coval's approach to evaluating voice AI performance is a direct descendant of work at Waymo building self-driving cars!
Voice agents are being deployed at scale for a growing variety of use cases. Today's LLMs are a solid foundation for building:
- customer support
- first-touch handling of telephone calls
- lead qualification
- scheduling and coordination
- copilots with access to specialized knowledge bases
Building products that work great in production is about more than just the LLM, though.
Testing/observability/eval tooling is a critical part of the emerging voice AI technology stack.
Announcing Coval's $3.3M seed round to build the evaluation infrastructure for reliable AI agents 🚀
Built on principles proven in autonomous vehicle testing, our platform enables engineering teams to rigorously evaluate and monitor AI agent performance at scale. From