Senior Software Engineer - Mercury Command

In 1965, an engineer in Scotland was given a mundane task with a tricky problem to it - banks wanted to close on Saturdays and still serve customers, but they didn’t know how to solve the authentication of the right user. James Goodfellow, working at Smiths Industries, uncovered the insights that you needed something you have (a card) and something you know (a PIN). Because someone told him they could only remember 4 digits instead of his proposed 6, today over 3 million ATMs use a card+4 digit PIN over 60 years later.

Like the ATM, Mercury is building technology that pushes forward financial interfaces for the long term. Command is Mercury's LLM-powered financial assistant, launched to all customers in June 2026. It lets users understand their finances and take action in plain language, from asking about cash flow to sending payments, issuing cards, and managing invoices. With the product now in customers' hands, the work is to evolve it, extend its capabilities, and find new ways to leverage LLMs to give Mercury customers a more powerful banking* experience.

What you'll do

Ship new capabilities users love:

• Design and ship new Command skills, the domain-specific instruction sets that teach the model how to handle workflows like sending money, managing invoices, and understanding cash flow

• Design and build agentic workflows in Command, defining the architecture for how multi-step agent interactions should work as we extend what the product can do on a customer's behalf

• Work with backend teams to define tool schemas for new capabilities, shaping the data contracts between Mercury's business logic and the model

• Own new capabilities end to end, from the system prompt to the frontend component that renders the response

Own the LLM layer:

• Maintain and evolve Command's prompt architecture: the system prompt, skill loading system, session context, and the polic

Department: Engineering