About the role#
This internship focuses on the security of Large Language Models and agent-based applications. You will join the Security AI team to research and build frameworks that protect our global products. This position is open to PhD students interested in the intersection of AI and security.
What you'll need#
- Currently pursuing a PhD in Computer Science or a related field.
- Deep knowledge of the theory and practice of LLMs and foundation models.
- A strong publication record at leading conferences.
- Excellent coding skills, including proficiency in Python, Go, or Java.
- Solid understanding of data structures and fundamental algorithms.
- Experience from internships, work, coding competitions, or publications in software engineering, NLP, or deep learning.
- Creativity and the ability to solve problems quickly.
- Strong communication skills and a collaborative mindset.
- Preference is given to winners of competitions such as ACM/ICPC, USACO, NOI, IOI, Top Coder, or Kaggle.
What you'll do#
- Research adversarial testing and evaluation technologies for foundation models and agents.
- Develop attack-defense detection methods, testing benchmarks, and toolchains for multimodal models.
- Create risk-defense technologies and runtime security assurance for agent-based applications.
- Work on training and inference technologies to ensure data and model asset security.
- Build and optimize confidential training and inference infrastructure.
- Research full-link defense systems based on trusted privacy computing.
Location & details#
- Location: San Jose, California.
- Modality: On-site.
- Employment: Full-time internship.
- Applications are reviewed on a rolling basis.


