AI Agent Engineer Intern

About this role

About the role: - Transform open-ended challenges in agentic AI into functioning prototypes, experiments, research outputs, or features that power the product. - Develop functioning agents from the ground up, establish evaluation methodologies, construct test environments, experiment with different planning approaches and memory systems, investigate why systems fail, and write up what you learn. - Work across research, engineering, and product—bringing perspectives from each discipline to your work. - Deliver outputs that range from production-ready modules and research publications to new benchmarks, internal utilities, and working prototypes. What we're looking for: - Fluency in Python and a track record of shipping code. - Direct experience building with or on top of large language models and AI systems. - Real curiosity about agents, evaluation frameworks, alignment, interpretability, enterprise-grade AI, or combinations of these. - Comfort in environments where the problem isn't fully defined—you scope it, build fast, absorb papers and documentation, write findings that others can learn from, and spot the highest-leverage next steps without being told. Bonus background: - Work with agent frameworks, systems that can plan over long sequences of steps, formal evaluation pipelines, or published research. - Contributions to or deep familiarity with open-source projects. - Backend engineering, enterprise data systems, AI safety, or model interpretability work.