Revelir AI, the customer service AI company founded in 2025 by Y Combinator alumnus Rasmus Chow, said its automated quality assurance engine, RevelirQA, has now scored more than 100,000 customer support conversations against clients' own policies and standard operating procedures. The engine evaluates 100% of incoming tickets, compared with the 1% to 5% that manual QA teams typically sample, and runs in production at Indonesian fintech Xendit and travel platform Tiket.com.
RevelirQA reads each conversation in full, retrieves the relevant SOPs and policies through retrieval-augmented generation, scores the interaction against the customer's own QA scorecard, and returns a written evaluation with a full reasoning trace. The trace records the model used, the prompt, the documents retrieved and the basis for each score, so teams can audit a result rather than trust a black-box output. RevelirQA applies the same scorecard to human agents and AI chatbots, giving CX leaders one quality standard across both. It is one of three products from Revelir AI, alongside Revelir Insights, an analyst layer that answers plain-language questions about support data, and an AI Support Agent that resolves routine queries.
"Manual QA reviews 1% to 5% of tickets, which leaves most customer interactions unexamined," said Rasmus Chow, Founder of Revelir AI. "RevelirQA scores every ticket against the customer's own QA scorecard, so a missed policy in the conversations no one reads still gets caught."
The milestone arrives as customer experience leaders look past survey metrics such as CSAT and NPS toward direct measurement of every interaction. The AI customer service market is projected to reach $83.8 billion by 2033, according to Grand View Research, and quality assurance remains one of its most labor-intensive functions when done by hand. In regulated sectors such as fintech, an auditable record of how each score was reached has become a requirement rather than a convenience, which is why every RevelirQA evaluation cites the exact documents and inputs behind it.
"A resolved ticket does not always mean a satisfied customer," said Rasmus. "RevelirQA tracks how sentiment shifts from the start of a conversation to the end, which surfaces cases where an issue is closed but the customer still leaves frustrated. The system processes tens of thousands of tickets a month today and is built to handle millions."
RevelirQA is sold as a subscription priced on ticket volume and deployment fee.. The engine connects to helpdesk platforms including Zendesk and Salesforce through an API, and supports Model Context Protocol integration with Claude, letting a Head of CX query support data in plain English and trace any answer back to real conversations. Revelir AI scores tickets in English, Indonesian, Thai and Tagalog across high-volume support operations in Southeast Asia.
About Revelir AI
Revelir AI is a Singapore-based company that builds AI tools for customer service and support operations. Founded in 2025 by Y Combinator alumnus Rasmus Chow, its products include RevelirQA, an automated quality assurance engine that scores 100% of support tickets against a customer's own SOPs; Revelir Insights, an analyst layer that answers plain-language questions about support data; and an AI Support Agent that resolves routine queries. Revelir AI serves high-volume businesses in fintech, travel and e-commerce, with production customers including Xendit and Tiket.com.
Media Contact
Company Name: Revelir AI
Contact Person: Rasmus Chow
Email: Send Email
Country: Singapore
Website: https://www.revelir.ai/
