Revelir AI launches AutoQA engine that scores every customer service conversation

Revelir AI has launched RevelirQA, an AutoQA engine that evaluates 100% of customer service conversations without manual sampling. The product is already running in production at Xendit and Tiket.com, aiming to give support, compliance, and product teams a full audit trail and a consistent quality standard.

Why it matters: - RevelirQA targets a core weakness in traditional QA: manual sampling that can miss most customer service issues. - The system scores every conversation against each company’s own SOPs and QA scorecards, which can surface hidden policy violations and recurring operational problems. - The product is built for companies that use both human agents and AI chatbots, where consistent quality measurement is harder to maintain.

What happened: - Revelir AI launched RevelirQA, an AutoQA engine that evaluates 100% of customer service conversations. - Xendit and Tiket.com are using RevelirQA in production, not just in pilot tests. - Both companies process thousands of customer service tickets per week, and RevelirQA scores the full volume against their existing QA standards. - Rasmus Chow founded Revelir AI in 2025. Chow is a Y Combinator W22 alumnus, and Revelir AI is headquartered in Singapore.

The details: - RevelirQA pulls each customer’s knowledge base and SOPs into a vector database using retrieval-augmented generation. - The system retrieves the relevant policy before scoring each conversation. - Every score includes an audit trail showing the model used, the documents retrieved, and the reasoning behind the result. - The platform is designed to compare support handled by humans and support handled by AI using the same standard. - Revelir AI says the system’s multilingual scoring has already been proven in high-volume environments, including Bahasa Indonesia. - RevelirQA is offered as a subscription SaaS product with three plans: Essential, Professional, and Enterprise. - Pricing depends on conversation volume and custom metrics. - Customers can deploy RevelirQA as standard SaaS or as a dedicated tenant. - The platform integrates through API with helpdesks including Zendesk and Salesforce. - Beyond QA scoring, RevelirQA adds signals such as sentiment, contact reason, and recurring issue type.

Between the lines: - The launch reflects a broader shift in Southeast Asia as companies move more customer interactions to chatbots while keeping human agents for complex cases. - Manual QA sampling typically reviews only 1% to 5% of tickets, which can leave most service problems invisible. - A system built around company-specific policies may be more useful than generic benchmarks for regulated sectors like fintech. - The audit trail is a key selling point for teams facing compliance review or regulator scrutiny.

What's next: - Revelir AI is positioning RevelirQA for enterprise buyers that need full-volume QA, custom policy scoring, and multilingual support. - The company also appears to be using its production deployments with Xendit and Tiket.com as proof points for broader enterprise adoption.

The bottom line: - RevelirQA is trying to replace sample-based QA with full-conversation scoring, giving support teams a more complete view of quality, compliance, and recurring customer issues.