Keri Warr
Implementing SL4/5 and searching for differentially defense-favored security tools.
Stream overview
Developing techniques and technologies to enable SL4 and SL5 cybersecurity postures for LLMs, such as hardware and software supply chain management, confidential computing, weight exfiltration prevention, ML compute cluster security, and AI-powered insider threat detection.
Mentors
Mentorship style
I love asynchronous collaboration and I'm happy to provide frequent small directional feedback, or do thorough reviews of your work with a bit more lead time. A typical week should look like either trying out a new angle on a problem, or making meaningful progress towards productionizing an existing approach.
Fellows we are looking for
Essential:
- Excited about doing Security Engineering work in particular
- Non-zero experience in all three of: Software engineering, Cybersecurity, and Machine Learning
- Ability to quickly unblock yourself
- Willing to trade off against how shiny your project is in favor of doing work that will be impactful within the next 12 months
Preferred:
- Significant depth in at least two of: Software engineering, Cybersecurity, or Machine Learning
- Excited about delivering working open source code in addition to research papers
Can independently find collaboraters, but not required.
Project selection
Mentor(s) will talk through project ideas with scholar, or scholar will pick from a list of projects.
The Winter 2026 cohort offers a wide range of research streams led by experts across AI alignment, interpretability, governance, and safety. Each stream provides its own research agenda, methodology, and mentorship focus.
