About this role
Summary: Designs and maintains evaluation frameworks and automated test harnesses that measure the accuracy and reliability of AI-powered accounting features. Provides engineers with tools and metrics to verify model performance before release.
About the Role
As a Member of Technical Staff - Evals at Entendre, you will play a key role in ensuring the quality and reliability of our AI-powered features. Your primary responsibilities will include:
- Designing and maintaining evaluation frameworks to measure the accuracy, reliability, and regression behavior of our AI capabilities.
- Building automated test harnesses that operate on real accounting data, identifying potential failures before they impact our customers.
- Defining metrics and benchmarks to provide the team with clear, quantitative insights into model and system performance.
- Creating easy-to-use tooling that enables engineers to write, execute, and interpret evaluation tasks as a seamless part of their development workflow.
- Collaborating closely with our Applied AI and Agent Engineering teams to guarantee every shipped capability has a well-defined standard of "good enough"—and a robust method to verify it.
What We're Looking For
- Experience writing and reviewing evaluation tasks used by AI researchers to assess and improve model performance.
- Background in quality assurance across software, client deliverables, or financial analysis.
- Exceptional written communication skills, with the ability to clearly articulate steps to achieve desired outcomes.
- (Preferred) Familiarity with accounting, financial systems, tax preparation, or payment technologies.