Senior Software Engineer

About the Role

We are hiring a senior engineer to lead technical strategy work and to build the proofs of concept that give that strategy its evidence. The role sits between engineering, architecture, and the business. In a typical quarter you will shape the architecture for a new class of system, get architects, risk partners, and product owners aligned on a path none of them owns alone, write the recommendation that goes to senior leadership, and build enough of the system to show that the recommendation holds. The work today centers on AI systems in regulated environments: agentic workflows and automations; the API, authorization, and fraud controls that make them safe; evaluation of systems whose outputs are non-deterministic; and the governance that lets them run in production. You will work with vendor platforms as often as with code we write ourselves, and much of the job is discovering what enterprise systems can actually do and what it would take to change them.

Responsibilities

• Develop technical strategy for the enterprise's highest-uncertainty technology questions: what to build, what to buy, what to wait on, and what has to be true for each path to work.
• Design reference architectures for new classes of systems, with particular attention to authorization, data access, audibility, and failure modes in a regulated environment.
• Demonstrate business value, feasibility, and scale in every recommendation, and say plainly when an idea should stop.
• Build coalitions across enterprise architecture, risk, fraud, information security, legal, product, and operations.
• Write the decision documents that let those groups agree, run the working sessions, and carry disagreement without stalling the work.
• Scope proofs of concept as hypotheses with a success measure and a stopping rule, then build them: working software against real enterprise systems and vendor platforms, built to answer a specific question and then handed off or retired.
• Lead the enterprise's approach to evaluating and enabling non-deterministic systems: test sets, assertion-based scoring, model-graded evaluation calibrated against human judgment, and the monitoring that follows a system into production.
• Evaluate vendor platforms and make build, buy