AI Engineer Intern

About this role

About the role: - Work with multiple AI decision models, connecting Claude and open-source options through a unified interface so different models can be used interchangeably. - Run comparative benchmarks to measure how various models perform against each other across different scenarios. - Enhance simulation systems built on Cloudflare Workers, Workflows, and D1 to make them faster and more reliable. - Add retry logic, output validation, tracking of data lineage, and spending controls to keep systems robust and cost-efficient. - Develop safeguards to catch when models produce made-up or nonsensical numerical results. - Set up monitoring and logging so that model predictions can be tracked and compared against what actually happened in the real world. What we're looking for: - You have completed at least one project where you deployed an LLM in production or shipped a feature using one. - You write production code in TypeScript or Python, with willingness to learn whichever is less familiar to you. - You think critically about model outputs and verify results rather than accepting them at face value. - You enjoy working across the AI and infrastructure stack with curiosity about how different pieces fit together.