About the role#
We are seeking an intern to help secure an enterprise AI capability built for senior executive leadership. Because this project involves highly sensitive information, identity, data protection, and AI-specific threat testing are core to the platform. You will work alongside Global Cybersecurity specialists and AI engineering interns to design security controls from the start.
What you'll need#
- Currently enrolled in a Bachelor's, Master's, or PhD program in Cybersecurity, Computer Science, Information Systems, or a related field with a Cybersecurity concentration.
- Strong grounding in identity and access management, encryption, logging, network security, and application security.
- Scripting ability in Python, PowerShell, or similar languages.
- Hands-on cybersecurity experience through labs, competitions such as CTFs, research, internships, or projects. Coursework alone is not sufficient.
- Hands-on experience with LLM applications, such as LLM APIs or frameworks, including a working understanding of RAG, embeddings, and how agents decide when to use tools.
- Awareness of LLM security and reliability risks, including hallucination, prompt injection through ingested content, and data leakage across access boundaries.
- Clear written and verbal communication skills, including the ability to explain technical work to non-technical and senior audiences.
- Curiosity, ownership, and comfort working with ambiguity in a highly regulated environment.
- Preferred: Experience with the OWASP Top 10 for LLM Applications or MITRE ATLAS, coursework in cyber ethics or law, and security certifications such as Security+.
What you'll do#
- Design need-to-know entitlements for executive-level and material non-public data.
- Support data classification, loss prevention, encryption, logging, and audit trails.
- Help review AI services and connectors against security requirements.
- Threat model an AI agent targeted at executives, covering prompt injection, data leakage, and misuse.
- Build and run tests for prompt injection, jailbreaks, and cross-boundary data leakage.
- Document red-team test results and recommend mitigations.
- Review connector configuration and help draft an incident response playbook.
- Advise AI engineering interns on guardrails and secure design patterns.
- Support audit and regulatory evidence for the governance seat.
- Help test secure data ingestion into Databricks.
Location & details#
- Term: Winter 2027 (January to June 2027).
- Commitment: At least 30 hours per week.
- Modality: Hybrid, with an onsite requirement of 4 days per week.
- Locations: Austin, Texas, or Quincy, Massachusetts.
- Employment: Full-time internship.
- Compensation: This is a paid position.


