Every new customer who applies for a Mercury bank* account is like a new star igniting in the night sky. At first glance, it's a single spark. But look closer, and a constellation emergesâidentity and business verification, ownership and control checks, sanctions and adverse media screening, and fraud signalsâall pulling on one another with invisible gravity.
The Risk Engineering team is the navigation system for this universe.
Our job is to guide thousands of ambitious people safely through Mercury's orbit every dayâeach with its own trajectory, risk profile, and story. We operate at escape velocity: moving fast, making confident decisions, and protecting the system without slowing down the builders who rely on us. When we do our work well, founders open bank accounts in delightful minutes, not daysâand never have to think about the complexity humming beneath the surface.
Now we're entering a new era. We're building agentic AI systems that don't just score risk, but reason about itâsystems that can investigate, decide, escalate, and learn. These agents will help power and automate our KYC Refresh, EDD, ODD, and compliance workflows, taking our operations to the next level of efficiency and scale. The goal: less manual toil, earlier risk detection, sharper decisions, and experiences that feel almost magical in their simplicity.
This is where you come in.
We're looking for a Senior Software Engineer who is energized by the idea of bringing intelligent automation to life in real production systems. Someone who wants to design and build AI-driven workflows that can think in steps, adapt to new information, and orchestrate complex decisions across people, models, and services. Someone who sees compliance not as red tape, but as a fascinating systems problem at planetary scale.
If you're excited to build agentic AI that mattersâsoftware that protects millions of dollars, unlocks opportunities for founders around the w
Please mention the word **CHASTE** and tag RMmEwMTo0ZmY6ZjA6ZWY1MDo6MQ== when applying to show you read the job post completely (#RMmEwMTo0ZmY6ZjA6ZWY1MDo6MQ==). This is a beta feature to avoid spam applicants. Companies can search these words to find applicants that read this and see they're human.