Gradient Labs is building AI agents that give every bank customer the experience of a dedicated account manager, addressing the complexity of banking customer service. The London-based company, founded by former Monzo AI leaders, is built on OpenAI models and is now shifting production traffic onto GPT-5.4 mini and nano. The platform achieves 500-millisecond latency for natural voice conversations.

In banking, customer interactions are governed by standard operating procedures that define what should happen at each step. A typical customer interaction might involve reporting a stolen card, with the system verifying identity, freezing the card, initiating a replacement, and answering follow-up questions. Each step follows a defined procedure with decisions made in real time based on user input and context. The company benchmarks providers on trajectory accuracy - whether the system follows the correct path from start to finish.

Gradient Labs' customers report CSAT scores as high as 98%, in some cases outperforming their best human agents. Most deployments start with over 50% resolution rates on day one, even for complex workflows like disputes and fraud. The company has increased revenue more than 10x over the past year, expanding from inbound support into outbound and back-office processes. Looking ahead, Gradient Labs is focused on systems that can carry context across interactions, understanding a customer's history and tracking ongoing issues.