AI agents are now common in customer service, sales, and product support. But as more companies turn to these systems to handle real conversations, a major problem has come into focus: no one can truly see what the agents are doing. That’s the blind spot InfiniteWatch aims to eliminate.
“Trillions of AI-powered interactions are coming — agents talking to customers, agents talking to agents. Companies won’t hand over critical operations to agents without visibility across QA, UX, compliance, revenue leakage, learning, reputation, and customer satisfaction. We’ve built the AI observability layer for this new reality. Traditional tools show you historical data. We predict what’s about to break and tell you how to fix it,” explains Pablo Molina, CEO and co-founder of InfiniteWatch.
The New York-based startup has come out of stealth with $4 million in pre-seed funding led by Base10 Partners, with additional backing from Sequoia and A16Z scouts, Kfund, Kibo, LifeX, and several unicorn founders. The company’s focus is on helping organisations observe, measure, and manage AI-driven interactions in real time: what its founders call the “agentic internet.”
Read More – Addition Therapeutics Raises $100M in Funding
Making AI systems as observable and trustworthy as any human-run process
The founders of InfiniteWatch, Pablo Molina and Iñaki Berenguer, were part of the leadership team that built CoverWallet, the insurance technology company acquired by Aon after scaling to more than $1 billion in premium revenue. They recognised a familiar gap forming in the AI world: fast-growing businesses adopting intelligent systems without the transparency needed to govern or improve them.
InfiniteWatch enables businesses to monitor and optimise how AI agents perform across voice and web interfaces. For teams already relying on AI for customer engagement, that visibility is essential to prevent minor issues from snowballing into lost revenue or degraded customer trust.
The company’s two initial products embody that vision. The Session Replay Agent reviews web interactions to automatically detect UX glitches and causes of conversion drops, pinpointing what’s blocking sales and recommending clear fixes. The Voice Agent plays both the tester and the auditor, creating synthetic personas to probe voice systems at scale and identify compliance and quality gaps before real customers experience them.
“When we deploy our observability solution, we can immediately offer to run the production AI voice agents for our customers, turning insights into live AI systems in a fraction of the time and cost of traditional approaches,” says Iñaki Berenguer, co-founder and President of InfiniteWatch.
Unlike web analytics tools such as FullStory, Hotjar, or Contentsquare, InfiniteWatch integrates monitoring and deployment into a single platform. That end-to-end approach, powered by a self-learning AI engine that processes millions of interactions each month, makes InfiniteWatch far more predictive than reactive.
Read More – Wearlinq Raises $14M in Series A Funding
