Machine Learning and AI Safety Research Intern

Apple

1d

About The Position

The Machine Learning and AI Safety Research Intern will conduct applied research at the intersection of generative AI safety, alignment, model robustness, and engineering. The intern will work with research and engineering teams to investigate AI safety challenges and turn findings into practical solutions that help protect users and deliver safer generative AI experiences. The role also includes developing evaluation resources and contributing to research papers. A 5- to 6-month internship between October 1, 2026, and September 30, 2027, is preferred.

Requirements

• Currently enrolled in an MS or Ph.D. program in Computer Science, Electrical Engineering, or a related quantitative field.
• Hands-on experience with LLM training frameworks such as DeepSpeed and machine learning ecosystems such as PyTorch and Hugging Face.
• Experience with LLM inference frameworks such as vLLM and programmatic LLM API calling, including client SDKs and batch evaluation techniques.
• Experience designing and building LLM evaluation pipelines, custom datasets, and benchmark suites.
• Proactive habit of staying up to date with emerging trends and state-of-the-art developments in generative AI and AI agents.
• Strong critical-thinking, problem-solving, and communication skills, including the ability to break down complex, ambiguous research problems into actionable engineering solutions.
• Deep familiarity with safety alignment concepts such as RLHF, guardrail implementation, and adversarial probing.

Nice To Haves

• Strong research experience in AI safety, model alignment, or red-teaming, ideally demonstrated by publications or papers under review at premier AI/ML academic venues.
• Preference for a 5- to 6-month internship duration between October 1, 2026, and September 30, 2027.

Responsibilities

• Conduct applied research on generative AI safety, model alignment, and robustness against adversarial attacks such as prompt injection, jailbreaks, and probing.
• Design, develop, and scale LLM evaluation pipelines, safety datasets, and benchmark suites to identify model vulnerabilities.
• Use and optimize LLM training frameworks and high-throughput inference engines for programmatic LLM API calling and batch evaluations.
• Collaborate with multidisciplinary research and engineering teams to translate safety concepts and guardrails into deployable engineering solutions.
• Synthesize experimental results and safety discoveries and co-author research papers for top-tier AI/ML conferences.

What This Job Offers

• Career Level: Intern