Coval
Simulation & Evaluation that scales voice and chat AI agents
Y CombinatorSummer 2024ActiveB2BEngineering, Product and DesignDeveloper ToolsSaaSB2BMonitoring
What they do
Problem
Teams building voice and chat AI agents struggle to reliably test them across many scenarios and catch regressions before deployment.
Solution
A simulation and CI/CD evaluation platform that automatically runs thousands of agent scenarios to measure performance and detect regressions across modalities.
Who it's for
AI engineers and product teams developing autonomous voice and chat assistants at startups and enterprises.
Founders
Brooke Hopkins
Founder
Building evaluation for AI agents, previously led eval infra at Waymo 🚙
Open roles 9
Company timeline
- Founded2024
- YC batchSummer 2024
- StatusActive
- Team size10 people
- LocationSan Francisco, CA, USA
Similar startups
DataSuite
Dataset Management for AI Trainers
Fulcrum
The agentic debugger for AI systems
PixelTable
Pixeltable is a declarative data infrastructure for building multimod…
Wato
Shared memory, tools, and workflows for agents across teams
camelAI
AI software engineer that lives on its own computer
Arena
LMArena began as a Berkeley research project and has quickly become e…