Mercury, a financial technology company, has been exploring the use of AI agents to enhance its operations and product development. Through internal hackathons and beta testing, staff engineer Matt Russell and his team have developed workflows that allow AI agents to autonomously handle routine tasks under human supervision, such as simple refactors and UI changes, thereby reducing tech debt and increasing team autonomy. The company has implemented AI for code suggestions and in-editor agentic chat, and is experimenting with integrating AI into user experiences, like natural language search capabilities. Despite the rapid advancements, Mercury maintains a high quality standard for its AI features, employing rigorous testing and gradual rollouts to ensure reliability. Russell envisions a future where senior engineers act as project leads, overseeing multiple AI agents to accelerate development processes, and advises continuous experimentation to stay abreast of AI advancements.