Machine Learning Engineering Intern

Zendesk is seeking a Machine Learning Engineering Intern to contribute to the Zendesk Explore platform. You will focus on building datasets and frameworks to evaluate AI agents, create "golden datasets" by annotating real conversations, and develop LLM-as-Judge evaluation logic. The role involves iterating on evaluation prompts and logic to ensure automated metrics accurately reflect human judgment and improve agent quality.

• Build datasets and frameworks for AI agent evaluation
• Create "golden datasets" by annotating real-world conversations
• Develop LLM-as-Judge evaluation logic using Python
• Iterate on evaluation prompts and logic based on human feedback
• Ensure automated metrics improve AI agent quality and behavior
• Work on both back-end and front-end of the Zendesk Explore platform