AI Researcher (Cybersecurity)

Company

hirify.global develops an AI penetration-testing platform with autonomous agents that find and validate exploitable cybersecurity risks for enterprise security teams. The company is backed by $51.5M in funding and has offices in Tel Aviv, the US, and the UK.

What you will do

• Research, train, and evaluate language models for cybersecurity and offensive security tasks.
• Design meaningful evaluations, analyze errors and failure modes, and distinguish real effects from noise and bugs.
• Develop post-training methods to improve model reasoning, tool use, exploration, and decision-making in challenging, long-horizon environments.
• Build large-scale data pipelines, reinforcement-learning environments, verification methods, and RL algorithms.
• Design, run, and analyze experiments, review relevant research, and implement new methods in Python using modern inference, training, and training-API tools such as vLLM, SGLang, verl, slime, and Tinker.
• Collaborate with cybersecurity researchers and infrastructure and engineering teams to scale research and translate outcomes into agent capabilities.

Requirements

• Master’s degree with a research thesis in Computer Science, Mathematics, Physics, Engineering, or another STEM field; or a Bachelor’s degree plus equivalent industry or army research experience.
• Hands-on experience in data-driven computational research, including machine learning, algorithm development, numerical analysis, statistical modeling, or similar work.
• Ability to take open-ended problems from vague ideas to working implementations.
• First-principles thinking and the ability to critically evaluate experimental data and distinguish real effects from noise and bugs.
• Strong understanding of deep learning and language models.
• Comfort with ambiguity and fast-paced research environments where priorities and core assumptions evolve quickly.
• Strong communication and collaboration skills for working in a small, agile team.
• Independence, curiosity, and a deep bias for action.

Nice to have

• PhD in a relevant field or relevant research publications.
• Background in cybersecurity.
• Background in reinforcement learning, particularly for language models or agents.
• Hands-on experience with deep-learning training or inference frameworks.
• Experience running and debugging large-scale distributed systems.

Culture & Benefits

• Work with a global team of offensive security experts, AI researchers, and engineers.
• Collaborate in a small, agile team in a fast-paced research environment.
• Work on cutting-edge AI and cybersecurity problems with direct impact on the product.