Revelir AI launches AutoQA system for customer service teams

Nov. 19, 2026
Revelir AI launches AutoQA system for customer service teams
By AI, Created 15:23 UTC, Nov 19, 2026, AGP -

Revelir AI introduced RevelirQA, an automated quality assurance platform that scores customer service conversations against company SOPs and QA scorecards. The Singapore-based startup says the system is already used by Xendit and Tiket.com and is calibrated to match human QA standards before deployment.

Why it matters: - Revelir AI is targeting a long-running support operations problem: manual QA typically reviews only 1% to 5% of customer service tickets. - The company is pitching full-coverage AutoQA as a way to catch repeated SOP breaches that sampling can miss. - RevelirQA is designed to give QA leaders a broader view of service quality across both human agents and chatbot interactions.

What happened: - Revelir AI launched RevelirQA, an AI quality assurance platform for customer service teams handling high volumes of conversations. - The company was founded in 2025 by Rasmus Chow and is headquartered in Singapore. - RevelirQA is already in use by Xendit and Tiket.com to evaluate thousands of tickets each week. - The platform is presented as a cloud-based, fully managed service that integrates with helpdesks including Zendesk and Salesforce.

The details: - RevelirQA evaluates each ticket against SOPs and a QA scorecard agreed with the customer’s team. - Before scoring, the system retrieves relevant SOPs and policies from the company knowledge base using retrieval-augmented generation, or RAG. - Each criteria score appears as Pass or Fail, with supporting conversation excerpts and the SOP document name used as the reference. - QA leads can override the score, and agents can dispute results if they believe the evaluation is wrong. - Revelir AI calibrates the system with each customer before launch. - The system is considered ready only after the customer’s QA analysts review a sample of results and confirm that more than 95 out of 100 evaluations are correct. - RevelirQA can flag errors such as incorrect refund amounts or missing regulatory information that agents should have given customers. - The platform applies the same QA scorecard to both chatbot AI and human agents. - RevelirQA also shows which customer issues are better handled by a chatbot versus a human agent. - Each score includes the model used, the prompt sent, the documents retrieved from the knowledge base, and the reasoning behind each scoring criterion. - Chow said the transparency is important for compliance needs at companies like Xendit. - RevelirQA does not automatically issue warnings or disciplinary action, and final decisions remain with the customer’s QA lead. - Revelir AI says the service can also surface customer sentiment, the reason for contact, and recurring issue types.

Between the lines: - Revelir AI is differentiating itself from broader AutoQA vendors by focusing on score accuracy and human calibration, not just coverage. - The company is also framing transparency as a compliance feature, which may matter for regulated or high-volume digital businesses. - The emphasis on sample representativeness suggests that volume alone is not enough if the sample set does not reflect everyday interactions. - A single dashboard for chatbot and human-agent QA could help teams compare where automation is working and where human intervention is still needed.

What's next: - Companies can test RevelirQA through a paid six-week pilot. - During the pilot, the AI evaluates 10,000 conversations based on definitions and accuracy targets set with the customer’s QA lead in week one. - The $10,000 pilot fee is fully credited toward deployment if the customer moves forward. - Revelir AI charges a one-time deployment fee for customer-specific customization, plus usage-based pricing tied to the number of conversations evaluated. - Revelir AI plans to keep positioning RevelirQA as a tool for CX leaders, support operations managers, and product leaders at digital companies with heavy customer interaction.

The bottom line: - Revelir AI is betting that enterprises will adopt AutoQA faster if the results are explainable, calibratable and close enough to human QA to replace most manual sampling.

Disclaimer: This article was produced by AGP Wire with the assistance of artificial intelligence based on original source content and has been refined to improve clarity, structure, and readability. This content is provided on an “as is” basis. While care has been taken in its preparation, it may contain inaccuracies or omissions, and readers should consult the original source and independently verify key information where appropriate. This content is for informational purposes only and does not constitute legal, financial, investment, or other professional advice.

Sign up for:

New Products Launch Guide

The daily local news briefing you can trust. Every day. Subscribe now.

By signing up, you agree to our Terms & Conditions.

Share this page:

Advanced Search Options

Search for:

Search scope:

Type:

Search in:

Date range:

The last

Sort by:

Sign up for:

New Products Launch Guide

The daily local news briefing you can trust. Every day. Subscribe now.

By signing up, you agree to our Terms & Conditions.