About the Role

We’re looking for a Model Interpretability Analyst to help our AI models make better decisions when evaluating candidate profiles. In this role, you’ll focus on reviewing model outputs, spotting inconsistencies or biases, and suggesting ways to improve how the model reasons. You don’t need deep technical or machine learning expertise—this position is all about careful attention to detail, clear communication, and strong collaborative skills to uncover and resolve issues.

If interested, please complete the trial here.

Key Responsibilities

• Review AI outputs to identify errors or patterns in how the model interprets data.
• Use an annotation platform to label and document these observations, ensuring information is organized and easy to share.
• Work closely with cross-functional teams (including product, data, and engineering) to discuss findings, propose improvements, and refine the model’s logic.
• Help develop a simple, organized way (taxonomy) to categorize different types of errors or issues.
• Suggest enhancements to workflows and tools to streamline the annotation process and improve overall accuracy.

Preferred Qualifications

• Strong analytical and critical thinking skills, with an eye for detail.
• Excellent communication and teamwork abilities—you’ll work with a variety of people from different backgrounds.
• Organized and methodical approach to managing large sets of data or information.
• Curiosity about AI and a willingness to learn; prior technical experience is not required.
• Comfort with trying new tools and processes to identify and address model errors more efficiently.

If you’re someone who loves digging into details, collaborating with others, and continuously looking for ways to make AI systems more insightful and fair, we’d love to hear from you!

‍