platoseedplatoseed
Test, monitor, and improve your voice agents
Roark helps teams test, monitor, and improve their voice agents. Voice AI is exploding, but reliability is still the #1 challenge. Teams spend hours manually testing agents and still miss failures. In the last 6 months, Roark has processed over 10M minutes of calls, powering monitoring and simulation for teams across the Voice AI ecosystem (including YC companies). We close that gap by combining: 📊 Monitoring & Evaluation - 40+ built-in metrics (latency, instruction-follow, sentiment, etc.), custom dashboards, alerts, and the ability to define your own. Up to 15 speakers with automatic speaker identification. 🎭 Simulations & Personas - end-to-end phone/WebSocket simulations for inbound and outbound agents, with configurable personas (accents, languages, speech & behavior profiles). Define tests as conversations with a graph-based approach, so you can easily branch into edge cases and variants. 🔁 Full lifecycle loop - failed calls automatically become repeatable tests, so every failure makes your agent stronger. Roark is the missing QA layer for Voice AI - helping teams ship agents they can trust.
Roark provides a Voice AI Testing & QA platform to monitor, evaluate, and simulate voice interactions. It helps teams catch issues before customers encounter them by monitoring calls, running simulations, and turning failures into tests.
The platform offers post-deployment monitoring with 40+ built-in metrics and multi-speaker analysis, live evaluators, and on-demand or automated evaluations via UI or SDK/API. It supports end-to-end simulations using graph-based conversation flows, configurable personas, and end-to-end testing for inbound/outbound agents. One-click native integrations (VAPI, Retell, LiveKit Cloud, Pipecat Cloud) and developer SDKs (Node, Python) enable quick setup and customization. Pricing is consumption-based with a minimum monthly spend, and the product includes dashboards, reports, and alerts with webhooks.
Who it’s for: Voice AI teams and organizations building or operating voice agents who need testing, monitoring, and QA for production calls.
Funding and accelerator backing noted (Y Combinator); active product with recorded usage metrics (10 million minutes of calls processed) and customer-facing demos, indicating early traction and ongoing development.
James Zammit (CEO) is an infrastructure and AI engineer with 10+ years of experience. Previously a Senior Engineer at AngelList, he built core portfolio infrastructure as the company grew from $10B to $124B in assets under management, and led the development of Relay, an AI-powered portfolio manager processing thousands of financial documents each month. He also co-founded a startup showcased at Google I/O 2016 in partnership with Firebase, and received scholarships from Meta and Apple.
Daniel Gauci (CTO) is a software engineer with 10+ years of experience. Previously a Senior Engineer at Akiflow (YC S20), he helped scale the company to $1.5M ARR and over 10,000 customers. Daniel spent seven years at Casumo, leading the development of a mobile app used by millions of players, significantly contributing to the company’s growth to over $50M ARR.
Replay real production calls against your latest agent changes, catch failures, and track sentiment.
Roark provides an observability and testing platform for Voice AI that replays real production calls against a updated agent to validate goals, monitor sentiment, and surface failures with dashboards and real-time alerts for teams shipping voice-enabled experiences.
Test, monitor, and improve your voice and chat AI agents
Complete QA platform for voice agents
