Senior AI Researcher
Core
Research LLM internals (activations, latent space, interpretability) to detect and prevent AI agent security attacks like jailbreaks and prompt injections.
Role type
Senior AI Researcher (LLM Security & Interpretability) (via careerplan.io/jobs/19-000-77-067-senior-ai-researcher-at-zenity)
Builds
Detection signals for AI agent security, open-source research tools, and tier-1 ML publications.
Domain
AI Security / Large Language Models / Machine Learning
Deliverable
production ML models
Required skills
Deep learning, LLM internals analysis, experimental design, scientific writing, data analysis, PyTorch, HuggingFace Transformers
Preferred skills
AI agent frameworks, RAG systems, production-grade codebase experience
Technologies
PyTorch, HuggingFace Transformers, LangChain, scikit-learn
Responsibilities
Investigate model internals via activation analysis and latent space discovery; Design security-grounded evaluations for attack types; Publish research in tier-1 venues (NeurIPS, ICML) and share open artifacts; Build and adapt tools/libraries for research automation.
Seniority
Senior, hands-on IC
