


Al-native customer operations for financial services


Gradient Labs is now a validated @googlecloud partner and available on @google Cloud Marketplace! We’re making it easier for banks and fintechs to deploy AI agents through their existing Google Cloud relationship, using the procurement and payment processes they already have. Dai Vu, Managing Director of Marketplace & ISV GTM Programs at Google Cloud, says: “Simplifying how enterprises procure and deploy critical technologies is a core component of accelerating digital transformation. By bringing its AI agents to Google Cloud Marketplace, @GradientLabsAI is providing customers with the trusted infrastructure and streamlined access needed to optimize operations and unlock real-world business value.” From frontline customer support to long-running back-office operations like KYC and disputes, we’re proud to partner with Google Cloud to power AI transformation in financial services. Find out more in the thread.

Our Backend Engineer, Vlad Tokarev, on what he’s observed lately when it comes to AI-generated code: lnkd.in/p/gUrk-yTi

Gradient Labs has placed 15th in @BCloudUK's FinTech 50 for 2026, the annual ranking of UK companies creating original technology for finance and payments.


A disputes cycle typically gets bogged down in the “waiting for evidence” stage. When a customer disputes a charge, cases often arrive to specialist teams incomplete, and it can take days for analysts to chase customers down for additional information. Automation seems like the obvious fix, but when UK fintech @yondercard first tried to automate their disputes cycle, they quickly hit a ceiling. The AI they tried initially wasn’t designed to carry context between the customer conversation and the back-office systems where disputes cases are decided. Even after early automation work, Yonder’s disputes cycle averaged 5–7 days, and that got longer during volume spikes. Then Yonder tried our Disputes Agent, which works across both sides of a case. The frontline agent takes the case and asks for evidence while the customer is still in the conversation. Meanwhile, the back-office agent classifies the case, checks the account history, and reaches back out in real time if anything is missing. Both agents work from the same case record and retain context, so the conversation flows seamlessly. After implementing our disputes agent, 80% of cases arrived to Yonder’s specialist team fully-evidenced. As a result, Yonder’s disputes cycle got 150% faster, and the average cycle stayed consistent even during volume spikes. AI that’s purpose-built for finance just works differently.

Find out more: cloud.google.com/find-a-part…

We’re excited to announce that @GradientLabsAI is now a validated @googlecloud partner and available on @google Cloud Marketplace. We’re making it easier for banks and fintechs to deploy AI agents through their existing Google Cloud relationship, using the procurement and payment processes they already have. From frontline customer support to long-running back-office operations like KYC and disputes, we’re proud to partner with Google Cloud to power AI transformation in financial services.

Most banks and fintechs have spent years building their customer-operations stack. Our Conversations API makes our AI agents headless, so the agent works inside the backend and channels you've already built. Our CTO @neal_lathia explains how: eu1.hubs.ly/H0zFlNZ0

We’re hiring an Events Marketing Manager to own our global events programme across the US, UK, and Europe, covering everything from intimate dinners to major conferences and partner events. Interested in jumping into a fast-paced team? Read more at our careers page: gradient-labs.ai/careers#ope…

Spotted: our logo on the big screen at @OpenAI's OpenAI’s DevDay Exchange, in a lineup of London builders we admire. Proud to be recognised with this cohort leading the way in the industry.


Watch our voice agent replace a stolen card in 5 languages. 🇬🇧🇫🇷🇧🇷🇩🇪🇲🇽 Every caller in this demo is someone from the Gradient Labs team (no actors), and this is how our agent responded to them. Learn more about our agent at the link in the comments. eu1.hubs.ly/H0zt3Vk0


If you’re running an AI agent at a fintech, you’ve probably encountered this headache already… so here’s how we’ve solved it. ⬇️

Busy phone lines mean your team likely doesn’t have the time to be proactive on customers’ accounts. If an application is incomplete or a balance is past due, that proactive call shouldn’t wait. Here’s how our Voice AI solves the bandwidth challenge for dozens of fintechs and banks: