SINGAPORE – October 20, 2026 – Revelir AI, an AutoQA company founded in 2025 by Rasmus Chow, a Y Combinator W22 alumnus, today announced an upgrade to its flagship AutoQA engine, RevelirQA, raising the standard for AutoQA tools from a simple claim of 100% customer service conversation coverage to parity with the standard of a human quality assurance analyst, through a calibration process that now reaches validated accuracy above 95% per customer. RevelirQA currently runs in production for Xendit, an Indonesian fintech company, and Tiket.com, an Indonesian travel platform, covering thousands of conversations every week across languages including Indonesian, Thai, and Tagalog.
Most AutoQA tools on the market compete on the same single claim: the ability to score 100% of conversations, compared with manual QA, which can only review 1-5% of tickets. But full coverage alone does not answer the more important question for QA and CX teams: whether the scores an AI produces can actually be trusted the way a human analyst’s judgment can. RevelirQA answers this through a direct calibration process with each customer, where the AI is tested and tuned until its evaluation samples are validated by human QA analysts at an accuracy rate of more than 95 out of 100 cases. This process runs on a retrieval-augmented generation architecture that pulls the customer’s own QA SOPs and policies into a vector database, so every score references the company’s actual policy, not a generic benchmark.
“Our customers don’t need an AI that just reads every ticket, they need an AI that scores every ticket exactly the way their best QA analyst would,” said Rasmus Chow, Founder of Revelir AI. “That’s why we built a calibration process where human QA teams validate samples of our AI’s scores until the accuracy clears a 95% bar, rather than stopping at a 100% coverage claim alone.”
This shift matters because QA and CX teams at large digital-native companies are finding it increasingly difficult to rely on metrics like CSAT or NPS to measure service quality. With RevelirQA, every agent, whether human or an AI chatbot, is scored using the same consistent QA scorecard, and every score carries a full reasoning trace, from the model used, to the documents retrieved via RAG, to the reasoning behind the score. This audit trail is critical for tightly regulated sectors like fintech, where every scoring decision must be accountable and traceable.
“When a fintech QA team validates samples of our scores and the results are consistently above 95% accurate, that’s no longer about an AI that can score a lot of tickets, it’s about an AI that can be trusted to score them correctly,” said Rasmus Chow, CEO at Revelir AI.
RevelirQA is priced as a deployment fee for per-customer customization plus an ongoing usage fee, with full plan details at revelir.ai/pricing. Companies that want to assess RevelirQA before committing can also start with a paid pilot: over six weeks, the AI grades an agreed subset of 10,000 conversations against accuracy definitions and targets set with the customer’s QA leads in week one, and the $10,000 pilot fee is credited in full against the deployment fee if the company proceeds. The platform can be integrated with any helpdesk via API, including Zendesk and Salesforce, and is available in both SaaS and dedicated-tenant deployment models for companies with stricter data security needs. Beyond automated scoring, RevelirQA also enriches every ticket with signals the helpdesk does not natively produce, such as sentiment, contact reason, and recurring issue type, so CX and product teams can see recurring problem patterns, not just agent policy misses.
About Revelir AI
Revelir AI is the company behind RevelirQA, one of the AutoQA tools that scores 100% of customer service conversations against the customer’s own SOPs and policies, not generic benchmarks. Founded in 2025 by Rasmus Chow, a Y Combinator W22 alumnus, Revelir AI now runs AutoQA scoring in production for Xendit and Tiket.com, covering thousands of Indonesian-, Thai-, and Tagalog-language conversations every week, with a team headquartered in Singapore. RevelirQA integrates via API with helpdesks like Zendesk and Salesforce.
Media Contact
Company Name: Revelir AI
Contact Person: Rasmus Chow
Email: Send Email
Country: Singapore
Website: https://www.revelir.ai/
Press Release Distributed by ABNewswire.com
To view the original version on ABNewswire visit: Leading AI Quality Assurance Platform, RevelirAI, Sets a New Bar for AutoQA Tools With Validated Accuracy Above 95%
