Design, prototype, and implement agentic workflows and AI-assisted tools for safety evaluations and operational-signal triage.
Test and characterize LLM and AI workflow capabilities, edge cases, hallucinations, and failure modes.
Integrate generative AI agents with structured safety databases and operational logs for real-time synthesis and risk reporting.
Collaborate with safety strategists and software engineers to test prototypes in operational scenarios and incorporate user feedback.
Develop evaluation harnesses and prototypes, APIs, or internal web dashboards for AI decision-support tools.
Requirements
Currently pursuing a master’s or advanced degree in Computer Science, Artificial Intelligence, Robotics, Data Science, or a related engineering field.
Must remain enrolled at an accredited college or university and maintain good academic standing throughout the internship.
Expected graduation between Fall 2027 and Summer 2028.
Strong software engineering fundamentals in Python and object-oriented programming.
Hands-on experience developing and deploying AI tools, LLM APIs, and agentic workflows.
Understanding of AI limitations, hallucinations, and verification methods for decision-making contexts.
Experience building working prototypes, APIs, or internal web dashboards.
Strong analytical problem-solving and written and verbal technical communication skills.
Bonus qualifications include applying AI/ML to autonomous systems, safety analytics, or risk management; designing generative AI and prompt evaluation harnesses; and familiarity with SQL/NoSQL database architectures and data warehousing platforms.
Benefits
Full-time, 12-week internship.
Relocation to the Bay Area, California, is required for the internship.
Interns must be enrolled throughout the internship and may not use Zoox proprietary information for an external thesis or publication.
Zoox
Zoox is transforming mobility-as-a-service by developing a fully autonomous, purpose-built fleet designed for AI to drive and humans to enjoy.