Solidroad Raises $25 Million to Scale AI QA

Solidroad announced a $25 million Series A led by Hedosophia, building on a $6.5 million seed from First Round Capital. The startup, founded in 2023 by Intercom alumni Mark Hughes and Patrick Finlay, operates from San Francisco and Dublin and runs an AI platform that evaluates 100% of customer interactions across voice, chat, and email to provide continuous quality assurance, coaching signals, and training simulations. Customers include Ryanair, Crypto.com, and Oura.

The platform automates QA by transcribing or ingesting conversation artifacts and scoring them against configurable company rubrics. Key practitioner-relevant capabilities include:

• •automated evaluation of 100% of interactions across human and AI agents
• •multi-channel ingestion: voice-to-text, chat logs, and email parsing
• •scoring and analytics that identify skill gaps, compliance risks, and coaching opportunities
• •generation of personalized training simulations for agents using conversation exemplars

Solidroad highlights large-scale throughput, claiming it has scored millions of interactions and improved analyst productivity by up to 10x. The stack details are not public, but the product positioning implies heavy use of conversation embeddings, classifier models for rubric mapping, and orchestration layers for routing insights into agent workflows and LMS systems.

Contact centers traditionally review 1-3% of interactions; Solidroad's value prop is statistical coverage at scale to convert conversations into reliable, actionable metrics. This addresses two industry pain points: QA is too sparse to be representative, and agent coaching is time-consuming to scale. The company's Intercom lineage and early enterprise logos give it domain credibility, and the funding signals investor appetite for tooling that augments human-in-the-loop support rather than replacing agents outright. As more enterprises deploy generative AI assistants, oversight tooling that evaluates both human and AI responses becomes a compliance and CX control point.

For ML engineers and CX leaders, Solidroad is an example of verticalized ML productization: combining transcription, intent/classification models, rubric-based scoring, and downstream simulation generation. Integrations with existing ticketing systems and contact center platforms will determine adoption velocity. Practitioners should evaluate such tools on evaluation coverage, rubric configurability, bias and fairness in scoring, and latency at scale.

Execution risks include classifier drift as customer language and product policies evolve, and the need for robust monitoring and human-in-the-loop feedback loops. Watch Solidroad's roadmap for richer explainability features, real-time QA hooks, and partnerships with major contact center platform vendors.

"We turn every customer interaction into measurable insight," said Solidroad co-founder and CEO Mark Hughes, framing the company as a QA platform meant to make exceptional service consistently achievable.

The Series A gives Solidroad runway to scale product and go-to-market, and it exemplifies a maturing market for AI-native QA tooling that complements generative assistants rather than competes with them. For teams building or buying QA systems, focus on configurable rubrics, drift mitigation, and seamless agent workflow integration.