Senior AI Researcher (AI)

TL;DR

Senior AI Researcher (LLM/AI Security): Researching LLM internals to improve the security of AI agents with an accent on interpretability, activation analysis, and attack detection. Focus on designing experiments on activations, uncovering mechanisms behind jailbreaks, and translating insights into detection signals.

Location: Tel Aviv

Company

hirify.global is a holistic platform built to secure and govern AI Agents from buildtime to runtime, recently raising $38 million in Series B funding.

What you will do

• Investigate model internals using activation/features analysis, unsupervised clustering, and discovery of directions in latent space.
• Design security-grounded evaluations and curate datasets for various attack types to evaluate white-box vs black-box methods.
• Produce research publications for tier-1 ML venues (NeurIPS, ICML) and security forums, and share open artifacts.
• Build and adapt research tools and agents to automate research work and distill knowledge from experiments.

Requirements

• Deep learning expertise with a track record of non-trivial research in LLMs or other domains (CV, speech).
• PhD or equivalent research experience in industry (5+ years in a leading research team).
• Strong experimental design and scientific writing skills, including comfort with testing causal claims.
• Proficiency with AI frameworks such as PyTorch, HuggingFace Transformers, and scikit-learn.
• Experience in data analysis, including visualization, exploration, and cleanup.

Nice to have

• Experience with a production-grade codebase with multiple contributors.
• Knowledge of GenAI tools such as LLM Orchestrations, Agents, and RAG systems.

Culture & Benefits

• Opportunity to work in a research-first role at the forefront of LLM interpretability.
• High-impact environment with funding to accelerate the mission of securing AI Agents everywhere.
• Support for publishing work in world-leading scientific venues.